Modern LED scrubber light

[rygh]

Enclosed are moderately detailed plans for an LED scrubber light.
This uses 5 state of the are high power, high efficiency LEDs. (3 warm white, 2 red)
It produces about 7,500 lux at 4" on a 6x9 scrubber. It uses about 15W.
Which I measure at a bit less light than 2 x 26W CFLs.
But should also have better wavelengths for plants.
It uses about 15W. So is considerably more efficient.
It costs about $70 to make. Cheap long term.
And everything near tank is only about 20V or so. Far less than dangerous 120V(600V) CFL.

It does produce a lot of light. Stay tuned for real Algae results.
I am worried it is too much.

PICTURES WILL BE INCLUDED BELOW

Light produced:
It produces about 7500 lux at 4 inches, over a 8x12 turf scrubber.
In comparison, two 26W CFL floods produce about 3100 lux over the same area.
* BIG Correction: CFLs produce about 9,900 lux over that area. Light was not warmed up enough during first measurement.
The wavelengths are plant-optimized, so very pink looking light.
It uses about 15W of power.
So it is about 3X as efficient, purely based on power/ lux measurements.
Given the wavelengths, that should go higher.
Picture of the lights on. Note that this was taken in a brightly lit room, but camera
had to drop the gain way down. It is much more pink/white than orange.
Digital camera has issues.

Components:
Purchased from ledsupply.com, for about $68.
Part numbers from their online store included.
2 x Luxeon Red K2 Star Low Current 05027-PD12
3 x Luxeon K2 TFFC Star 140 Lumen 05027-PWW4-140
1 x Xitanium: Xitanium 17W 700mA Off-line Driver XI-LED120A0700C24F
1 x XIitanium Driver Connector 1365323-1
Arctic Silver thermal epoxy.
2”x2” x 3’ aluminum angle. Thin.
9” x 12” x ¾” plywood
6”x12” thin plexiglass.
Power cord, small red and black wires, screws, misc.

STEP 1: Build board and heat sinks/brackets:
Cut board to size. Mine was 9x12, but match your ATS.
If you have a router and 45-degree bit:
Drill holes slightly larger than LED stars.
Route out around the holes.
Otherwise, just cut larger holes for the LED stars.
Cut the 2x2 aluminum angle iron into 6” strips.
Rough on outer side, for gluing later.
My holes were a bit close together, so I offset them. No real need for that.
Then paint. Best is OUTDOOR, flat white.
Now, screw the angle brackets to the back of the board, vertically.
Rough side facing through holes.
Now (Critical)
Label them. Left->Right. Label board as well.
Trace a circle onto aluminum where you can see it through holes.
Drill 2 small holes for wires, in aluminum, about ¾” apart, at very edge of hole.
Now, remove angle brackets from board.

STEP 2: Build LEDs on brackets.
First, solder red wire to “+”, and black wire to “-“, on LED stars.
Does not matter which side.
Then, insert wires through small holes you drilled in aluminum bracket.
Find small clamps / clothespins.
You should be able to clamp LED stars inside the small circle you traced.
Wires should be inside as well.
Check it with board.
If all is well, mix up the thermal epoxy, and glue them all down.
Clamp very loosely. Only use a thin layer of epoxy.
Steps 1+2 can be done in one day. Let paint an epoxy dry.

STEP 3: Connect it all up.
Screw angle brackets back onto board. Be careful of LEDs.
Then wire it all up.
“Plus” from power supply -> 4 foot long wire -> Red(+) of first LED.
Black(-) of first LED to Red(+) of second LED.
Connect Black to Red for the rest.
Final Black(-) of last LED -> 4 foot long wire -> “Minus” on power supply.
Then connect power cord to power supply line side.
Fat plug is neutral. Thinner one is line.
The 4 foot long wire is optional. It simply allows you to put power supply and all 120V things
well away from the tank.

Test is quickly.
If all is well, I suggest soldering all connections.

STEP 4: splash guard.
Cut a thin piece of acrylic to cover front.
This is somewhat optional, if there is no splash.
I added small ¼” standoffs, to keep acrylic from heating up, but that was really not needed at all.

[rygh]

Pictures to help explain steps.

[rygh]

More pics

[johnrt]

I am really looking forward to watching this develop.

I wish I had something constructive to add, but great project.

John T

[kcress]

Fore shore! Nice so far!

[ocean rock]

, need to see a few more pic , are the led rated for 10,000 hrs use before , need replacing , ?

[kcress]

HBLEDS are usually rated at 50,000 hrs not 10,000.

Of course if he's really going to run them at 1A he will likely not archive that number.

[rygh]

They are rated to 50,000 hours at 1000mA. I am running them at 750mA.
My expectation is at least 5 years. Probably 10 if I push it. By then, I will probably want to try something different anyway.
BTW: The other very key component to LED longevity is heat.
If you note, each LED has its own 6" aluminum heat sink, and ends up only slightly warm to the touch.
In my experience of failed LED lights, it is ALWAYS the power supply that dies first.
I did use what I hope is a good supply. We shall see, especially with a bit of salty humidity.
Although I did put the power supply outside of the sump cabinet.

[rygh]

Ocean Rock : Any particular pictures or description you might want.
I just covered the basics. Easy to add more.
Well, except for a picture facing the LEDs, with them on. Completely blinded the camera, and me.

[kcress]

rygh; Is that second picture the back of the fully ASSEMBLED light?

[johnrt]

"Any particular pictures or description you might want."

I think we all want to see green goo. So do you.

John T

[rygh]

Yes.

[inkidu]

In my experience of failed LED lights, it is ALWAYS the power supply that dies first.
I did use what I hope is a good supply. We shall see, especially with a bit of salty humidity.
Great setup thanks for the ideas. I am trying to figure out leds and I noticed that the mean well led driver I am going to use, LPLC-18-700, has an IP or Ingress Protection # With this particular led driver it is IP 67 which means you can immerse in water up to 1m. Something I will never try to do but good to know if there is a freak accident, like dropping it into the tank, you might have some protection. Trying to figure out the best way to waterproof wire connections to make as safe a setup as possible. Any ideas? Here is a link for the other IP codes
http://www.taitmobilebop.co.nz/IP67
Hope this helps. Note not all mean well led drivers have the same rating and I still have to use the one I am talking about.

[rygh]

Well, simply use aquarium safe silicone sealant on all the external wires. You probably have that lying around already.
It should seal well, is non conductive.
However, for right around the LED itself, I am not so sure. Could be thermal issues.
If you need to seal there, use a non conductive epoxy, like they use on PCB boards.
But safety is really not much of an issue. It is only 16V or so. If dropped in the main tank, might freak out the fish,
but should not kill them. Would definitely not kill you.
As long as you keep the 120VAC power supply well away of course.

[rygh]

Clarification:
What I did was have a very short 120V wire, from plug to power supply, and then a long (4 foot) wire from
power supply to LEDs. The long wire runs low voltage. With the short 120V wire, no real way to drop it in the tank.
But be a little careful of heat buildup on the power supply itself as well. Generated a few watts. So no fan needed,
but cannot really be completely enclosed in a small box.

[kcress]

Bump for answer.

[rygh]

Yes, second image is the fully assembled light.
My previous yes was too vague, and confused with a different post.
Note that the wire going out at top right is the 4' one going to power supply.
Third (front) image shows power supply.

[rygh]

UPDATE:
Well, I am seeing a bit real algae growth now. Finally!!
But not much yet. Hmm. I will add pics soon.
As background, my problem is that I have a really good ATS established in the main tank.
Otherwise known as annoying hair algae all over the place.
My nitrates and phosphates have always been undetectable.
So the point here is more about algae removal (moving) than reducing nitrates. Which according to forums, is harder.
Note that I also have a good protein skimmer (still running), and remote deep sand bed.
With the previous CFL lighting, even after 3 weeks, all I ever had was a pale yellow.
Plus a bit of oily-green crud grew up near the pipe, where lights barely hit.
Now:
I am getting a bit of hairy green here and there. Encouraging, but slow.
Seems to be growing best on the outside, where there is less light. Possibly too intense. We shall see.
I also added a lawnmower blenny, who seems to be eating a bit.

[SantaMonica]

With the previous CFL lighting, even after 3 weeks, all I ever had was a pale yellow.

Because your bulbs were to near, for the amount of flow it had.


Plus a bit of oily-green crud grew up near the pipe, where lights barely hit.

Because that was far enough away from the bulb to not get over-lit. The oily stuff is what you get when your nutrients are high.

[rygh]

Interesting.
I am probably way over-lighting with the LEDs then. I will move them back a ways, and see what happens.

I don't really think water flow can be an issue though.
It is single sided, and probably has around 60 gph / inch. (Only a 10" wide scrubber, with entire sump flow)
Way more than 35 gph. May it be too much? Although I keep reading you can never have too much.
The thin-green-slime area was also up where there would be less flow, not just less light.

[SantaMonica]

You'll find that beam-spread is the problem with LED's. Not spectrum. It's an optical issue probably best solved by hundreds of smaller sources, with a diffuser.

[kcress]

Try some of that diamond diffuser?

Thanks for the clarification.

When you do pics I'd like to see a pulled-back shot that shows the lighting if you can manage.

[rygh]

Beam spread should not be an issue with this build.
The raw LED has a micro lens on it, with a "Lambertian" radiation pattern.
I am not adding any extra optics. With narrow beam optics, it would definitely be a problem.
You have to look in the datasheet for a detailed drawing, but basically, the radiation pattern is pretty spread out.
Relatively shallow curve between +/- 15 degrees.
Not steep really until +/-40 degrees.
So with 5 overlapping sources, coverage is fairly even.
Sources are about 2" apart, where target is 5" away. So a +/-45 degree angle is a lot of overlap.
Sure, not as diffuse/good as a big wide CFL bulb, but far better than a single metal halide.
But the concern is worth checking. I can move the light meter around the target area, and see what the difference is.
Nothing like real measurements.
That, and take some pictures. But I've been pretty busy lately.

Diffuser plate would certainly spread things out even more, but also absorb a lot of your light.
Even the thin plexiglass I used made a noticeable difference.

[rygh]

Ok, some more detailed measurements on spread / light distribution:
On original distance, lux range min/max over target distance = 5500 - 9800. (1.8X)
At 6 inches, lux range = 2,700 - 4,800. (1.7X)
Error in measurement is fairly large. Probably 20%+. Surprisingly hard to get a solid measurement.
That may seem like a problem, BUT:
Dual CFL setup at 6" = 5,500 - 10,000 (1.8X)
So yes, there are light distribution issues, but really not worse than dual CFL.
* However, I would expect a long tube CFL to be a lot better.

[rygh]

"Before" picture of tank.
Not bad, but plenty of short hair algae.

[rygh]

My LED / turf scrubber in action.

[rygh]

Nice comparison of lights.
15 Watt LED on the left.
2 x 26W = 54 Watt Normal dual CFL on the right.
LED is clearly pinker, but not quite as bright. Glare makes it a bit deceptive.
When I originally set it up, I was more excited, since LED seemed brighter.
But I made a mistake in not letting CFL bulbs warm up enough.

[rygh]

My ATS screen. Almost exactly two weeks.
I have cleaned it a small bit 3 times.
A lot better than before, where it was simply pale yellow, even after 3 weeks.
More on conclusions soon.

[rygh]

MY OVERALL CONCLUSIONS (OPINIONS) SO FAR:
Did the LED lighting improve my growth?
- Clearly Yes. Time will tell if it is truly good.
- BUT: I am quite certain I could have fixed it with repositioning/improving a normal CFL design as well.
- So do NOT do this project to "fix" your ATS. Do it for fun!
Does it use less power?
- Clearly yes.
- The light is slightly dimmer, but can be positioned closer easily.
- You can tune the spectrum.
- It is only 15 Watts, instead of 54.
Was it hard?
- Well, much harder than a simple CFL bulb in a home-depot socket.
- But on par with doing your own T5/Ballast setup. (Which I have also done on main tank)
Was it expensive?
- Well, yes. But only initial cost. Over time, should be cheaper , with no bulbs to replace, and power savings.
Other issues:
- Could be a beam spread problem. But after measuring, I think I must disagree with our resident expert on that.
- Spread is not constant. Far from perfect. But really no worse than a normal CFL setup. A T5 would be better though.
- It takes up less space. Produces less heat.
- Lower voltage near tank is MUCH safer. Did you know CFL starting voltages can be more than 400V!
- No mercury. And with changing bulbs that often, not and entirely trivial issue.

I am looking forward to the long term of seeing if the whole ATS concept will improve my tank.

[kcress]

Thanks for the pics! And the run down.

[SantaMonica]

Thing is, if that pic is after two weeks of LED, and if that is an improvement over your CFL's, then your CFL's were unfortunately not operating correctly in the first place. The CFL screen should have been packed with algae, at least 1/2 inch thick, after two weeks (of course you normally would not let it go two weeks). Also notice the circles on screen, where the pods are eating

[rygh]

My theory on slow growth, largely from reading your FAQ here:
* Competition for nutrients from algae in main tank.
--- My Nitrates and Phosphates tested as zero.
--- The algae in the tank is very well established. And not truly rampantly growing like early on, just tough slow growing tufts.
--- I also have a protein skimmer, tons of live rock, and a remote deep sand bed.
* My ATS design is not the best.
--- It is single sided, not entirely vertical, and has only one layer of screen.
--- I did the hole-saw thing, but probably not enough.
* The CFL setup I had was probably wrong. Too close/far or something.
So the combination of all that originally allowed the main tank algae to completely out compete the ATS algae.
Now, I am just getting slow growth because of first two issues above.

However, I am curious:
Your last post seems to imply that you expect the CFLs to work far better than the LEDs.
I really do not understand why that would be the case. Measurements show them as very similar. Both produce plenty of algae-usable light.
I do plan to try diamond diffuser plate, but again, measurements show similar issues with edison socket CFLs.
The new-ish high powered LEDs are well proven for greenhouses and main aquarium tanks. My main tank is about 50/50 T5/LED.
I would expect LEDs to be comparable in effectiveness, but better for environmental/other reasons.

BTW: Don't take a little bit of argument here as not appreciating this great site.

[SantaMonica]

There has not been an LED scrubber that has worked, yet. So that's why I say a CFL should work better. Display lighting does not apply, since your eyes can't tell the difference in power levels. But algae can, and over-lit areas will yellow, and under-lit areas won't grow, even though the LED coverage looks the same to your eyes. I do think LED's scrubbers are the future, but for now they are impossible to DIY.

As for your CFL screen not growing, you are right that's it's a combination of things which made your scrubber too weak to get started. I would tell most people to (even if temporarily) double or triple their scrubber power. I think the singe-sided part of yours is what did it in.

[drhyne]

What about these new LED panels they are selling all over eBay. They range from 14w to 90w. Everything from solid colors to quads. The quads are mostly Red and Orange but fill with Blue and Full Spectrum White. I have a neighbor that grows "herbs" in his basement with these. He can't keep up with the rate of growth on his plants.

Features & Benefits: Item Includes:
Each light has 90 x 1w LEDs, 6:1:1:1 red/blue/orange/white
Precision red & orange LEDs from High Power Opto
Precision blue & white LEDs from Bridgelux USA
Orange & white LEDs provide fill for all missing spectrums
True optimized spectrum with tight-tolerance LEDs
Unrivaled chlorophyll/carotenes/xanthophylls absorption
Uses wide-angle (120°) LEDs for max coverage & output
Extremely efficient, integral reflector
Dazzling, powerful light
No infrared rays or ultraviolet radiation to stunt plants
LEDs emit light without using filaments; light runs cool
Built-in, ballast-free, power supply
10-years or more life-expectancy with minimum decay
Three, thermostatically-controlled, low-noise fans
Operates on 110 or 220 volts; standard 3-prong cord included
(we will supply free adaptor for any UK buyers)
Built-in lugs for easy mounting & hanging
CE & ROHS compliant
Brand new in original box
Absolutely the finest lights available at any price
Risk free purchase; 30-day money back guarantee

1 - 90-watt Xen-lux light
Carabiner hanging kit
6' grounded, 110v USA cord
(UK buyers receive adaptor)


Specifications:
Number of lights: 1
LEDs per light: 90
Total LEDs (1-watt): 90
Color: Red/blue/orange/white
Input current: 0.8 amps
Power per light: 90w
Unit diameter: 10.75"
Unit height: 2.5"
Frame: Metal
Red wavelength: 660 nm
Blue wavelength: 455 nm
Orange wavelength: 610 nm
White wave: full-spectrum / wide-kelvin
Max coverage: 290 sq. ft.
Use: Indoor growing
Shipping weight: 8 lbs.

[kcress]

Go ahead. We'll watch over your shoulder. :D

[drhyne]

.... gotta wait for my tax return!

[rygh]

I cannot find the detailed specs for those LEDs.
But if they are high efficiency, like Cree/Luxeon, 90W would be BLAZINGLY bright.
Could be equivalent to 120W+ of CFL lighting, so likely way overkill for an ATS.
The usual 14W panels I have seen use the tiny LEDs. Those are FAR less efficient, so 14W would be very little.
But I don't know exactly which ones you are looking at.
Make sure you look at the detailed specs before buying!

[rygh]

I am trying to figure out leds and I noticed that the mean well led driver I am going to use, LPLC-18-700, has an IP or Ingress Protection
I looked at that spec in more detail, and looks like a great product.
In fact, I bought two of the 35W versions for a different LED project. Unrelated to fish.

[rygh]

Quick update:
I added 2 more layers to the screen, and they are MUCH rougher. Stay tuned next week on that.
To be scientific, I only added them on the lower part, so I can sortof compare.

There is a "trick" to the "hole saw trick". (SM: If you see this, you may want to try it, and add it to the FAQ if it works well)
Do not simply scrape the hole saw laying flat around the screen.
Tilt it slightly, such that the teeth of the saw are cutting like they normally would in a board.
When pulling toward you, tilt left side up. When pushing away, tilt right side up.
While a bit tricky, it worked a lot better. Less work, much rougher.
I think flat, the teeth angled the other way actually push the plastic back down, wasting effort, and breaking off what was pulled up.

Also, my flow is not what I thought. I double checked, and it is more like 40 GPH. But single sided, so well above the requirements.

[SantaMonica]

Yes I use that tilting trick.

[inkidu]

Does anyone have an opinion about the importance of CRI for growing algae?


The color rendering index (CRI) (sometimes called color rendition index), is a quantitative measure of the ability of a light source to reproduce the colors of various objects faithfully in comparison with an ideal or natural light source. Thanks wikipedia


From what I have been reading I notice that some bulbs that are classified as full spectrum are done so because they have a high CRI i.e. greater than 80 and also a color temp higher than 5000K

Although I have a seen grow bulbs with only CRI values greater than 60. I guess with a grow bulb you are less likely to be concerned with how the colors of the plant appear and might have different mix of kinds of phosphor to achieve the different wave lengths that would be beneficial.

It seems that there is no real standard for terms like full spectrum and grow bulbs and they may not be interchangeable.

Why I ask is there seems to be a trend for led makers to produce leds with higher CRI values.

Namely I am looking at The LXML-PW71 Luxeon® Rebel White - ANSI 3000K which has a 90 CRI and a very narrow 2870K-3220K color temp

Sharp also has a new line of 6.7 watt Zenigata LEDs that have a high CRI (87) which only started to be manufactured last month.

Thanks for any help

[rygh]

You do not really want full spectrum for growing algae.
Well, specifically, it is a waste of power. It does not actually hurt.
Chlorophyll (Algae) are green, because they do not use the green part of the spectrum.
Or for that matter, a whole lot of the center.
Look at en.wikipedia.org/wiki/File:Chlorofilab.png for a nice drawing of the spectrum that plants like.
You really want something high in blue/purple and red/orange.

In a perfect world, you could pick exactly what you need.
An easy alternative, is "warm white" with a bit of extra red.
The nice thing about "warm white" LEDs, is that they are fairly close to what you need, but wide spectrum enough that
you don't have to be very exact..
I added some red for the upper end. I have some red/orange on order, to see what that looks like as well.
The fundamental color of those white LEDs is really blue. They use phosphors, activated by the blue, to produce the other colors.
Look at LED datasheets for spectrums produced.

[rygh]

By the way, do not equate narrow color temp with narrow spectrum.
Narrow color temp simply means that it does not change much from batch to batch, component to component, and current to current.
Really more of a measurement of quality.

For that matter, color temp does always tell you much about true spectrum anyway.
It is simply the average of the spectrum. What YOU perceive. Well, sort-of. Simple way to think about it.
Since algae ignore so much of the spectrum, it can be very different from what they perceive.
And LEDs have such a strange spectrum, the average can even be in a wavelength that they barely produce.

[inkidu]

Thanks for the reply

Why I ask is I am getting growth on my horizontal algae scrubber but not too much, granted it has only been going for less than a week. I am impatient

because I

bought 6 discus 5 of which are really sick from LFS. They weren't having to much success so I picked up all of them for $100. A steal if they live.

I added 4 normal 2700K 13 watt cf two days ago and 5 rebel cool white + 1 red rebel too the scrubber today. I have a feeling that since the colored leds have a

narrow frequency that it would be easier to dial in white leds if you have the right kind. Possibly ones with a high CRI. There seems to be new kinds of leds

coming out every month. Understandably you will be wasting some frequencies of light but with all the other benefits of leds I will look past that to get results.

Has anyone have an opinion on all the 10,20,30,50, and 100 watt leds that are all over ebay? 100 watt warm white led with a diffuser might work well.

It could be put together for $!50 and that is 6000 lm. I am going 1o add a 10 watt warm white and a 10 watt red once they arrive.

[rygh]

.... I have a feeling that since the colored leds have a narrow frequency that it would be easier to dial in white leds if you have the right kind. ..... Understandably you will be wasting some frequencies of light but with all the other benefits of leds I will look past that to get results. ....
Yes, this was similar to my thinking as well. Get something spread out, then fill whatever spectrum holes are left.
I have 3 warm white, and 2 red.
But looking at the waveform, I am wondering if I need more blue.
There is a real gap between 450 nm and 500 nm.
The blue part of the while LEDs seems to be more in the violet range, and very tight.

KEY PROBLEM with the big high wattage LEDs is the heat sink!
At 2W, a simple aluminum angle bracket is fine.
At 50W+, you are talking a CPU quality heat sink, with fan.

Another issue SM already pointed out is diffusion of the light. Fixable with diffusers, but that absorbs light.

I might very well experiment with more pure LEDs though.
I am thinking purple + blue + red-orange + red, all in equal numbers. Probably 3 of each.

[rygh]

GENERAL UPDATE:
My ATS is still not growing well at all. Minimal slow growth.
That really says little about LED/CFL btw. My old CFL was worse. But could have been placed wrong.

That said, I am fairly happy with so far anyway.
The hair algae in the main tank is actually looking a bit less healthy.
It was not really growing before, but not dying either. Now it is going a bit grey where light is less intense.
My nitrates remain at zero, so nothing expected there anyway.
My corals seem a bit happier. Especially my pipe-organ. This is probably due to the increase in pods, or
maybe a subtle increase in water quality.

NEXT STEP:
I first have to modify my sump. Various reasons. But part of that will be to lower it, allowing more room for ATS.
Then, I plan to make an improved two sided ATS.
At that point, I will use standard CFL on one side, and LED on the other.
Real scientific like.
Also, it will then be large enough that I can more safely experiment with turning off my protein skimmer.

[SantaMonica]

I really think the answer is an array of low power LED's; one every 1/2 inch or so. Then of course there is the light-screen version:

viewtopic.php?f=3&t=34 (http://www.algaescrubber.net/forums/viewtopic.php?f=3&t=34)

[rygh]

That really only solves the diffusion issue.
And I did measure it, and I just don't see how that could really be a problem.
But even if it is, it can be solved pretty easily with a simple diffusion screen.

You need to be really careful with efficiency.
- I assume when you say "low power LEDs", you mean the modern "bright" 5mm ones. Those can be good or bad.
- But the old tiny ones are terrible as far as cost/lum and watts/lum. Don't even go there.
- Power supply becomes an issue as well. You either need really high voltage for the long chain, or several small supplies.
- Small supplies are usually not PWM/DC-DC, so far less efficient.

The big problem of course, is that it is a royal pain to build.
- That is a lot of soldering to do.

But yes, ignoring effort, with a proper design, and especially with a real PCB, it might be the best.

[inkidu]

I got 4 of these heat sinks for $64 shipped off of ebay.

Aluminum Heat Sink 10"x7"x3"

I connected six rebels and have not notice the heat sink even get warm

Probably over kill but I wanted to run fan less

Do you think a heat sink that size could handle a 100 watt led?

I have some experience wiring 120 mm computer fans to run at 5 volts making them nearly silent.

I have a local glass store in town that showed me a completely clear glass with a wavy surface.

Might still produce hot spots but maybe less of them. At least it would not block transmission so much.

Anyone try something like that? Do you think it might be an good diffuser?

I trying to get large wattages because it seems cheaper and easier to setup than so many small ones.

What does anyone think of a 100 watt warm with some distance between the screen and a diffuser?

Thanks for any help

[kcress]

Well.. You can clearly over-watt algae. I don't know at what level though.

That size heat sink would be enough for 100W especially if you have a fan moving the air correctly past the fins.

You should probably keep the light back far enough to have relatively uniform light across the screen.

More sane would probably be a handful of 1W LEDs spread around heatsink. That way you get more uniform coverage without the concentrated heat.

[inkidu]

Thanks for the reply

"More sane would probably be a handful of 1W LEDs spread around heatsink."

More sane maybe but also a lot more work and to some extent cost per lumen.

I have yet to install but the 10 watt warm led I just got is blinding. (same construction as the 100watt) I am curious to see what a light meter would read.

Could possibly go with something not so strong (50 watt) ???

Could anyone give an opinion about how far a 100 watt or 50 watt should be place away from a screen?

Would not some distance solve some problems with the light being to intense in any one area?

[kcress]

Move it back until it covers the whole screen and try it.

[rygh]

I got 4 of these heat sinks for $64 shipped off of ebay. Aluminum Heat Sink 10"x7"x3"
Do you think a heat sink that size could handle a 100 watt led?
Probably.
A quick check on a standard 9x6x2 heat sink = 0.5 degC/Watt without fan.
Unfortunately, I don't know the exact efficiency of Light/Heat for those big LEDs. Lets guess at a fairly standard
70% heat, 30% light.
I also don't know the package thermal characteristics. Lets guess at 0.75 degC/Watt.
So you need to dissipate 70W. 70*1.25 = 87 degrees C.
Assuming you have 80 degF ambient around fishtank, that is 26 deg C.
So you are at 26 + 87 = 113
You absolutely need to keep them below 135, or they start to fail, so some margin there.
But you will lose a lot of efficiency and longevity as things heat up.
I strongly suggest a fan.

Also note: In researching this, I found two really interesting things of note.
1) Due to the way they are built and thermal issues, expect a 50% reduction in efficiency from a single LED.
Remember, they are specified cold, but running very hot.
2) The output light is not as dispersed, and very intense. It can cause cloth to burn and plastic to melt!

[SantaMonica]

Consider a water-cooled version

[inkidu]

Thanks for the reply. Well I just hooked up the 10 watt red and 10 watt warm to one of the heat sinks that had 6 rebels on them and after a few hours touching

the integrated heatsink of the led its seems to be pretty cool although my house is fairly cool. Kind of subjective but I am guessing something like body

temperature. Thats making me happy as well I am now starting to see a fair amount of algae growing after 12 days. I wish my camera was not broken because

there seems to be quite assortment of different algae ( some green and brown hair) Mainly because starting from my inlet the water passes under a 10

watt warm 10 watt red, then a cluster of 5 rebel cool white and one red, and then some common phillips 2700 k cf which gradually transition to a mixed line of

3 rebel red, 2 rebel blues and a rebel cool white. Brighter green under the cool whites transitioning to darker brown towards the end. Some of the results so

far might also be effected by the flow i.e. approx. 400 gph over 6" wide and 36" length horizontal screen. The flow seems more turbulent in the beginning. I am

also missing the baffles that came with the scrubber that increased the flow through the screen. I am working on that. Still a way to go to determine what will

work with my setup but I feeling better that I am starting to see results. Eventually I will add a fan I agree with rygh the heat sink will work many times better

with one. I am running out of places to plug things in. As far as efficiency I am trying to do some math too. I am far from believing all the limited specs from

ebay but this is what is I get from the 100 watt warm white led. (The way I am trying to compare efficiency is not conventional or maybe

not even right because I use forward voltage multiplied by current.)



forward voltage of 36 * 3.5 amps = 126 watts which supposedly put out 6000 lumen

6000 lumen / 126 watts = 47.6 lumen per watt

As a comparison a rebel warm typical forward voltage 3.4 * .7 amps = 2.38 watts which puts

out 110 lumen at .7 amps

110 / 2.38 = 46.2 lumen per watt

another comparison a cree warm forward voltage 3.4 * .7 amps = 2.38 watts which puts out 130 lumen ( my approx. its 93.9 at 350 mA)

130/2.38 = 54.6 lumen per watt

There must be a different way of calculating these #'s because I do see things like 160 lumen per watt efficiency for a cree ( the cool white are brighter 122

lumen compared to 93.9 at .35 amps but nothing that would make that much difference.)

I must be comparing something different. Like I said this is just how I am comparing them. After a little reading, wilkpedia, they do seem to use a much different equation.

http://upload.wikimedia.org/math/8/d/d/ ... ebf141.png (http://upload.wikimedia.org/math/8/d/d/8dd17c7c28522e4b602c8f3865ebf141.png)

Not that its terrible important seems that this could be a fair way to compare different leds. Not sure????

power in light out

Anyone weigh in?

[rygh]

Yes, lumens/watt is a VERY good way of getting efficiency. As is lumens/dollar.
And you seem to be calculating it right.
I think you used the 350mA luminance values for 700mA current in one spot though. Not sure.

Cree Q5, XR-E Star, Cool white:
107 lm@350 mA
3.3Vf @ 350 mA
So power = Current * Voltage = 350mA * 3.3V = 1.155W
Thus power efficiency of Lum/W = 107 lm / 1.155W = 92.64 lm/W

Now to save cost, you will probably end up running at 700mA, which does drops it down a bit.
And they are a bit sneaky in not reporting all 700mA numbers in their data sheet.
But it is pretty close to double, and Vf only goes from 3.3 to 3.5.

[inkidu]

Thanks for the reply I agree rygh Cree for some reason does do some odd things with there specs.

This is some more info I found. Its validity like any thing on the internet should be ? here is some more math

(not warm white they produce less lumen)

- True Cree Q5 bin emitter
- Cree Flux/Color BIN: Q5 WG
- 3.7V typical driving voltage
- Manufacturer Rated 228 Lumens at 1000mA input current:
350mA: 107~114lm
700mA: 171.2~182.4lm
1000mA: 214 ~ 228lm

3.5 * .7 = 2.45 watt

176 / 2.45 = 71.8 lumen per watt
avg.

3.7 * 1000 mA = 3.7 watt

221 / 3.7 = 59.7 lumen per watt
avg.

As you increase your current(amps) through the led the forward voltage increases ( thats why I used 3.5 for .7 and 3.7 for 1 amp

although the specs don't clearly state what current the 3.7 forward voltage pertains too. (must likely 1000 mA 3.7 is pretty high)

There is some variability to begin with. Usually given as a min, typ, amd max #'s. I personally find rebels specs a little better.


Something I want to point out, as you try to run leds like a rebel or cree xr-e at higher currents they start to be less efficient.

There is also a dramatic loss of over all life with the higher currents and the corresponding temps.

So instead of essential overvolting these would not undervolting a 10, 20, 50, or even a 100watt led make more sense?

I am far from an expert and I would appreciate anyones opinions.

[rygh]

Cree Vf:
350mA=3.3V
700mA=3.5V
1000mA=3.7V

Yes, efficiency drops significantly the higher the drive.
BTW: There is an extra factor that is not in the datasheet.
- The higher you drive it, the hotter it gets. The hotter it gets, the lower the output. But lumens are all specd at the same temp.

So driving at 350mA versus 700mA is far better. --- except for that pesky issue of ending up at double the
cost due to purchasing twice as many LEDs.

Datasheets I usually use: ledsupply.com/docs/cree-xre.pdf

On the 100W LEDs: These are not really a single 100W diode. They are an array of diodes. Maybe 7x7 each at 2W. Not sure on detail.
There is an interesting failure mode that they don't mention much.
In an array, in row dimension, they depend on the LEDs being all about the same, and sharing current evenly.
As they deteriorate over time, the LEDs start changing Vf a bit, and current is less shared.
That can end up over driving other rows.
Also, if any cell completely dies, that row stops drawing current, and that current goes into the other rows.
You can get a cascade effect.

At any rate, it can be analyzed endlessly. But then, analyzing it is the fun part.
- Light efficiency = Lumens/Watt. Highest Q, lower current to maximize.
- Cost efficiency = Lumens/Dollars. Upper Q, higher current to maximize.
- Overall build hassle: Generally the fewer the LEDs the better, until fancy heat sinks required.
- Longevity: Top brand name, keep temperature way down, keep current well below max.
- Spectrum control: More LEDs mean you can pick and choose colors. (Affects "useful" lumens)

[inkidu]

Thanks for the reply

I count 10 x 10

I am thinking there must be some cost savings because the manufacture doesn't need to cut them all up

and can keep them all in one package. I assume the downside would be you get some diodes that are better than others.

Such is life. The good with the bad.

I might buy a higher watt one once I can see how the 10 watt led I have does. I figure its worth a try.

Thanks for the help.

[ChrisD]

Inkidu

Make sure you look up the chromacity group (e.g. WG, WC) on the cree datasheets. The binning on the leds can be a very big range - 5000K to 10000K from memory. One of the graphs in the datasheet shows where these sit if you know the grouping. I'm not qualified to comment on how this may affect performance in a scrubber or main tank but it should certainly affect aesthetics just as a 5000k tube would look substantially different to a 10K one.

edit: If you are looking for efficiency and are prepared to do the work to connect up multiple leds you may find some deals on the cree xp-e, xp-c or xp-g chips. Get them pre-mounted as they are tiny (about 3mm square). Some hit more than 130 lumens per watt and I have seen pre-mounted verrsions giving abouto 70 lumens per euro in terms of cost efficiency.

[whites]

This is really taking fish keeping to a new level...... I like

just a thought with red and blue, is it true that blue is only needed if you are trying to flower or fruit a plant so dose that mean algea needs blue to flower or fruit? in which case we only need red or is that the other way around?

And a qustion at what ranges should red and blue be at?? should it be full red range and full blue range? I ask as I am looking at getting some GU10 light fittings to play with 1x red 1x blue they are 3w and pretty cheap. Just to see how much quicker I can grow algea under a coloured light nm specific range.

I have just a small plastic canvas setup up with 2 panels each panel is lit by a 3w GU10 globe warm white 4000k and seems to be doing pretty well for the size. the lens are a little to focused tho I think they are 20*.
A week and a half in and this is what they are like.. one panel has been in longer the one on the right is older by about 6 days, both have had a scraping (very little came off) the count on the canvas is too low I need to get some #14 as the water dosent stay on the canvas that well oh and I didnt seed them either. With higher wattage they should do well then ill make a bigger one after the experiments of led when these screens are mature and move the skimmer on.

http://i694.photobucket.com/albums/vv310/shane_photos_video/IMG_3791.jpg

http://i694.photobucket.com/albums/vv310/shane_photos_video/IMG_3792.jpg

http://i694.photobucket.com/albums/vv310/shane_photos_video/IMG_3793.jpg

http://i694.photobucket.com/albums/vv310/shane_photos_video/IMG_3789.jpg

[SantaMonica]

Reminder, LED's are for experimenting only, not results.

You don't need to seed. If it's going to grow, the screen will be covered thick in 7 days.

[rygh]

Reminder, LED's are for experimenting only, not results.

You don't need to seed. If it's going to grow, the screen will be covered thick in 7 days.

Some points:
1) You are getting pretty negative on LEDs. They ARE the future, as soon as people figure out how to implement them correctly.
Until then, expect a lot of failures. Sound like ATS history?

2) From reading the forums, there are PLENTY of non-led failures still, even when done according to faq.
So never make quick assumptions about root cause.

3) You may in fact need to seed.
Kcress seems to have proven that in his build thread.

[inkidu]

I think I might have found a path to the future or at least something interesting.

http://www.mouser.com/ProductDetail/Led ... oVIu8lRQxj (http://www.mouser.com/ProductDetail/LedEngin/LZC-70WW40/?qs=sGAEpiMZZMvdqMP/aZ%252bGCuoVIu8lRQxj)

If you check the specs they seem very good ( a lot of high wavelength red and some low blue )

From what I have researched so far these two conditions are not in many leds.

By the way the part of my scrubber that is only lit by leds is working.

Still thinking about going vertical.

They also have a 10 watt 661 nm deep red led (expensive $30)

[SantaMonica]

Until then, expect a lot of failures

Yes, and I don't want the people that come here looking to fix their tanks to think that an LED will do the fixing, because then they'll say they tried it and that scrubbers don't work.


there are PLENTY of non-led failures still

There is not one. Show me if I missed it.


You may in fact need to seed. Kcress seems to have proven that in his build thread.

You made my point. Thousands of people have grown thick gha in the first 7 days of a screen's life, without seeding. Hundreds who have tried seeding did not get any better growth, but did end up with a spike from the dying of the seed. Thus, seeding is not recommended if the goal is to fix the tank. I have over 100 builds that I have not posted yet, and none of them are seeded.

[rygh]

I think I might have found a path to the future or at least something interesting.
http://www.mouser.com/ProductDetail/Led ... oVIu8lRQxj (http://www.mouser.com/ProductDetail/LedEngin/LZC-70WW40/?qs=sGAEpiMZZMvdqMP/aZ%252bGCuoVIu8lRQxj)
If you check the specs they seem very good ( a lot of high wavelength red and some low blue )

Interesting.
Spectrum looks fairly similar to Cree "Neutral white", but possibly a bit better, yes.
It is 2000lm/40W = 50 lm/W. Not bad, but not great either.

Hey: If you build a vertical, it might be fun to do LED on one side, and CFL on the other.
Real scientific comparison that way.

[whites]

email was sent to me... my quest continues to find out what band of red we really need as this email states its for flowering and fruiting. It seems we can find out heaps of what the algea needs to feed and flow etc but light is still a problem to pin point.

I know you guys are going lower kelvin cfls to get some red in there or is it orange that we are after??

Can anyone point me in the right direction to try and figue this one out.

Dear Mr.Shane Grant,

Thank you for your prompt reply!

Here I send the PI of quad band 50w led grow light for you as attached file.

All of our price are in $US.

Blue light is good for vegetatives and the red light is good for flowering and budding, you can choose the ratio of the light, which have wonderful efficacy for the plant.

Vegetative(Dual-band): Red: 620~630nm / Blue: 450~465nm R/B ratio: 8:1 or 7:2
Flowering(Dual-band): Red: 660nm / Blue: 450~465nm R/B ratio: 8:1 or 7:2
Vegetative(Tri-band): Red: 620~630nm / Orange: 610~615nm / Blue: 450~465nm R/O/B ratio: 7:1:1
Flowering (Tri-band) : Red: 660nm / Orange: 610~615nm / Blue: 450~465nm R/O/B ratio: 7:1:1
Vegetative(Dual-band): Red: 620~630nm / Orange: 610~615nm / Blue: 450~465nm/White R/O/B/W ratio: 6:1:1:1
Flowering (Dual-band) : Red: 660nm / Orange: 610~615nm / Blue: 450~465nm/White R/O/B/W ratio: 6:1:1:1

Yours sincerely,
Michael


I was heading in the quad band vegative direction what do you think??

[SantaMonica]

These might help


http://www.radio-media.com/fish/ChlorophyllSpectrum.png

http://www.radio-media.com/fish/ChlorophyllSpectrum2.jpg

http://www.radio-media.com/fish/ChlorophyllSpectrum3.jpg

[inkidu]

I just bought 6 of these 10 watt deep red 661 nm leds ( wavelength is in the zone for chlorophyll A)

http://www.mouser.com/ProductDetail/Led ... 7STztIY%3d (http://www.mouser.com/ProductDetail/LedEngin/LZ4-40R210/?qs=sGAEpiMZZMsgllGlynFdfmaQ2jWZYDUYN7YB7STztIY%3d )

Combining those with 6 of these 5 watt warm white ( they have a particular high red wavelengths and low blue )

http://www.mouser.com/ProductDetail/Led ... tHhobz8%3d (http://www.mouser.com/ProductDetail/LedEngin/LZ1-10WW05/?qs=sGAEpiMZZMt82OzCyDsLFJq/dm5mXjJxGCpgtHhobz8%3d)

I am hoping the mix will let the different alga be satisfied through the natural progression of the screen.

[rygh]

Just a note (copied from web):
Within the algae there are 4 types of chlorophyll a, b, c (c1 and c2) and d.
Chlorophyll a is found in all photosynthetic algae, whereas chlorophyll b is confined to Euglenophyta and Chlorophyta.
Chlorophyll c is found in Dinophyta, Cryptophyta, Rhaphidophyta, Bacillariophyta, Chrysophyta, Xanthophyta, Phaeophyta and Prymnesiophyta.
Lastly, chlorophyll d is a minor component of many red algae.

I think green hair algae is Chlorophyta Derbesia.
So a mixture of A+B is good.
However, it might even be a win to emphasize the B spectrum, to help it out compete algae that only has A.

[inkidu]

Thanks for the info

Rygh I have seen some info about chlorophyll c and d but never anything about

the spectrum they like. ????

[rygh]

I could not find any waveforms on that either.
But found this:

Carotenoids and chlorophyll b absorb some of the energy in the green wavelength. Why not so much in the orange and yellow wavelengths?
Both chlorophylls also absorb in the orange-red end of the spectrum (with longer wavelengths and lower energy). The origins of photosynthetic organisms in the sea may account for this. Shorter wavelengths (with more energy) do not penetrate much below 5 meters deep in sea water. The ability to absorb some energy from the longer (hence more penetrating) wavelengths might have been an advantage to early photosynthetic algae that were not able to be in the upper (photic) zone of the sea all the time.

[whites]

GREAT INFO!!


based on what your saying which one do you think I should go for?

Its all abit confusing to me. I kind of understand but not really lol.

based on the graphs blue seems to be a bigger range including purple? and orange is incorperated as well with low reds being good to with a high spike in mid blue and mid red.

My options are dual, tri + quad and in flowering or vegative.

still my thoughts are to go quad and vegative? maybe add some purple, deleting the white for purple and half white/purple, they can make what I want. with 50w at my disposal :)

Any thoughts guys

[rygh]

Confusing to all of us.

I would suggest the "flowering", because the vegetive skips 660nm, (chlorophyll-A red)
Beyond that, not sure.
Seem like ideal is to cover 400-500, 620, 660.
With the ratios they are quoting, it is almost all red.
Another issue is LED accuracy. The actual spectrum on an LED can vary quite a bit from "standard advertised".

[rygh]

So browsing last night, I noticed a company doing something with algae bio-oil.
(Sorry, forgot to save the link)
They said they did a lot of research on lighting, and they were using LEDs for lighting.
This was for free-floating algae, not what we want, but close.
I noticed a 4-1 ratio, 4 red for every 1 blue.
Impossible to tell exact spectrum of course.

[rygh]

UPDATE ON MY BUILD:
Well, I am still fighting the dark-crud. I am cleaning every 3 days, but the crud still seems to be winning.
The problem is, I just built a small test ATS, not the real deal.
It is effectively 30 sq in, single sided, for a 65G fairly loaded well fed tank.
So it cannot keep up.
Plus, there is a flow problem. My actual flow is a LOT less than I expected from the pump rating.

The corals do seem a bit happier though. More open. My leather has been growing like a weed. And I notice more tiny pods in the water.
Nitrates/phosphates still read 0, always have.
But still plenty of short-hair algae in the main tank.
So for what it does, I would say it is a partial success.
Enough to spend the effort to build a full size one. Unfortunately, with a separate pump.
Someday. This is Valentines day weekend. "Honey, I need to spend all day in the garage building something to grow more algae". Hmm. No.

[inkidu]

I notice you did not have the prices in the list you posted. Why I mention it from my research so far

leds are very expensive when the tech is pushed to produce wavelengths toward uv (blue) or infrared (red)

So I would be suspect if there is not a considerable increase in cost when high red (660) is being used

(EX. of the reds I purchased there is increase of over 20% from normal red, 630nm, to deep red, 660nm)

especially considering that there is more red leds than any other color ( and I have seen higher cost differences)

I also noticed that as time goes by there is an increase in multi band led panels ( ironic that the dope smokers

are doing so much to increase the experimental knowledge about led grow lights) which include whites. I assume

it is because there is still a need for a broad spectrum.

Hope this helps.

[rygh]

I do not see a cost difference where I bought mine. ledsupply.com
All the colors, and non-sorted white, are the same price.
Only the high lumen specially sorted whites are more.
Both luxeon and cree.
That includes the royal-blue, but they do not have a deep red.
Also, while pretty good pricing, not exactly a major wholesaler type.

[inkidu]

rygh, your comments lead me to what I think. to some extent, is the crux of the problem.

The normal blues, even royal blue, might not be low enough but even more importantly the

reds are not high enough.

Ex a rebel royal blue (440nm to 460nm) or a Cree royal blue(450 to 465) still are a little off of what is needed

for chlorophyll a (also keeping in mind that the led is predominately in the middle region of of those ranges i.e. has a sharp peak)

Also more importantly, for at least the reason that there are on several orders of magnitude more reds, that

most red leds are in the low 600nm range. This also misses to some extent the peak red levels of chlorophyll a.

The deep red I have on order is 661nm. I am far from an expert and this is only how I am interpreting the info I have seen so far.

hope this helps

I noticed that the temp that leds run at can shift the wavelength produced. Anyone know in which direction?

[rygh]

Interesting. Maybe that is why the normal LEDs do not succeed so well.
Interesting product: http://www.ledengin.com/products/5wLZ/LZ1-00R205.pdf
High power, deep red, specifically for clorphyll-A

[whites]

Cost to my door was $109au, $94us something like that pretty close to it anyway this was the best I could come up with, Others were the same price for 37w and a 45w, the diffrent lights stay the same price from this particular company. I guess that they make so many of the damn things they can make it cheap. The cost of delivery is almost the same as the unit 2.5kg. Unit on its own was $62us + $32us del.

If it dosent grow the algea then I can grow tomatoes ;)

So what was your choice inkidu? I thought the same about full range and vegative seemed to pinpoint spec a little.

Thanks for the thoughts guys btw its been a big help making things clearer. My drama is I really wanted to get as close to the spec as possible for the hobbys sake but I still think a fuller spec is the way to go. including the orange I sure the plant would use this as well in its healthy growth.

[inkidu]

You sure beat me on price. $83 on rebels, $36 on 3 drivers, and $30 on two 10 watt prolights

with $221 in ledengin leds and $85 in meanwell drivers on the way.

I hope that there long life pays off.

[whites]

pluss your time making them thats at least $50 p/h lol

[SantaMonica]

I still think a useful design would be a large number of low power led's, spread evenly across the screen (1W X 200). Just do some wave soldering. And use the water for cooling, by extending the ground grid up into the flow, through the sealing resin.

Also, I'll re-post my earlier idea of a light-screen:

http://www.radio-media.com/fish/LightScreenDrawing.jpg



A light-screen is a scrubber where the screen IS the light, and the light IS the screen. Compared to regular scrubbers, these plastic-covered LED light screens:

o Are ultra small/thin.
o Have no algae die-off (see drawing below).
o Are practically unbreakable.
o Are electrically safe (12 volts or less).
o Can be made as small as desired for nano's.
o Can easily be built into the hood of a nano.
o Are double-sided with almost no increase in size.


Disadvantages:

o They will be expensive (equivalent to good skimmers).
o They are impossible to DIY



http://www.radio-media.com/fish/LightScreenAdvantages.jpg



Here is my version of a nano scrubber:

http://www.radio-media.com/fish/Nano.jpg



Same concept, just smaller, and replaces the skimmer, mechanical filter, and other filtering "devices" in pre-fab nano's like Aquapods, Red Sea Max's, etc. Would actually make nano's less expensive, better filtered, more compact, and more reliable.
.
.

[rygh]

Something along those lines might be a fiber optic version.

Use large 10W+ LEDs, so reasonably easy to DIY.
Each LED drives the end of a small packet of SIDE GLOW fiber optic cables.
Basically, a bundle of small cables.
Now each of those cables you spread out, and simply weave into the standard plastic mesh.
The "side glow" fiber optic cables let out a bit of light on the entire length, unlike standard fiber optic cables that
purely transmit from one end to the other.
Something like 6 bundles at 10W each, 15 small fiber optic cables for each bundle.
So 90 little light pipes, about every 1/8 inch.

Another thought is to find a broken LCD TV that has the new LED backlighting.
On those, there is a fancy glass backing, where LEDs drive at the sides, and are spread out very evenly over the front.
A bit like a freznel lens.
Maybe you can even buy replacement glass for the back.

[SantaMonica]

I like the side-glow idea. A mesh of these in resin could be either a light-screen, or a light source for a regular screen. No heat, no electricity.

[whites]

Any links to this side glow led fiber optic cable??

[rygh]

Just an example:
http://www.thefiberopticstore.com/purchase/sideglow.htm
You could take the 84 strand one, and maybe strip the PVC jacket off of part of it.

Another idea would be "LED rope lights".
These are a large series of small LEDs, already connected together.
http://www.affordablequalitylighting.co ... index.html (http://www.affordablequalitylighting.com/docs/indoor/rope/led/led/index.html)
I think there was some industrial version as well, with simply LEDs mounted on a flexible PCB.

But I really have three problems with all of these ideas:
1) Efficiency. Smaller LEDs are rarely as efficient. Light is lost in various resins and such. Definitely a big loss in fiber optic.
Basically, a huge point of LED is to beat CFL in efficiency. I hate to cut into that win.
2) Light up from the roots seems backwards.
My impression was that hair algae grows simplistically like a tree, with a root of sorts on one end, and major growth on the other.
So it seems that light from the screen itself would be blocked by the older already grown algae, and not hit the new tips.
3) All of the heat goes into the tank.
Ok, not huge, but still, if lighting is part of the screen, 100% of the heat will be transferred to the water.
Of course, conversely, no need for a heat sink.

[SantaMonica]

Light up from the roots seems backwards. My impression was that hair algae grows simplistically like a tree, with a root of sorts on one end, and major growth on the other.

Algae just grows where it has nutrients and light. No real roots. So a light-screen will reduce growth as it gets thicker (and farther away from the light).

The major, huge, gigantic advantage of a light-screen is that it can be left unattended, forever. Well, in theory at least; it has not been tried yet. Regular screens must be cleaned weekly because new growth smothers the old. But a light-screen stops growing once it's full, and then just sits there. Therefore there is no new growth to smother the old. True, the filtering will slow down and stop too until cleaned, but the advantage is still a huge one for new users who many not know (or want, or be able) to clean weekly.

True about the heat.

[rygh]

Has anyone tried the simpler version of this:
A single sided design, with a solid clear backing on the mesh, and put the lights on the other side.
Making sure no water is on the light side.
Basically, light it conventionally, but from screen side.

[SantaMonica]

No but it could work. Led or otherwise. With led's, it would be like the screen I posted, but one-sided led's, and laying down.

You could almost think a light-screen as flypaper; as soon as it fills up, there is no more "stickyness'" for more algae.

[whites]

So would a one sided screen in those dimentions work for a tank of say 300L? or would it need to be bigger?

Then what wattage would we need? Ive seen 15w to aobut 25W in the flat panel style.

Diffusion of light would just be a matter of some plastic fluro lighting gear like what you see in supermarkets.

My experiments with 3w white led seem that the angle of lense is to great as the light spot where its focused is too bright. I have to move them far away from the screen say about 15-20cm to get a larger light spread. You can see this on the pics I posted. Just a thought when playing with the light screen Idea.

Santa your light screen would be cheaper to make too I would think once the design gets tweaked.

[SantaMonica]

If you are asking me, I have no idea what size or wattage would be needed for anything. It has not been done. One thing, it would be weaker that a regular scrubber, because of the growth-stoppage.

[whites]

I know its not led but I thought this was interesting.

http://www.parans.com/

[inkidu]

Take a look at these freakish things sst-90 red (forward voltage 2.2 at 6.3 amps 800 lm)

http://www.luminus.com/stuff/contentmgr ... nation.pdf (http://www.luminus.com/stuff/contentmgr/files/0/d2efe3cc2679a82099c0e5e4843b4469/miscdocs/pds_001421_rev_02_sst_90_rgb_product_datasheet__il lumination.pdf)

I found them on the avnet site,never ordered from, for $33. What I think might be useful

is that they bin for wavelength. rygh wrote "However, it might even be a win to emphasize the B spectrum,

to help it out compete algae that only has A." So maybe you can better dial in exactly what you need.

I have found out that you can't always find out what bin # ( mouser for ex. ) unless you buy a 1000/a lot.

I am also coming to realize that the absorption spectra can change. Trying to figure this out now.

Just trying suggest some options I notice it does not have near the lifetime of the deep reds I got from ledengin.

Hope this helps.

[rygh]

So I am pretty sure what I will do.
I have been busy constructing a single sided, 6.5" wide x 30"long (35" with ends) next generation ATS.
I will use 2 sets of LED lights. One at first, for experimenting, and comparing with CFL.
Each set:
- 4 deep red, LedEngin
- 4 red or red/orange K2s
- 2 blue K2s
- 2 royal blue K2s
So a ratio of 2:1 red/blue. And deep red/royal blue, for some fun spectrum coverage.
The blue puts out a lot less light per watt, so real light ratio is pretty red.
All running at 700mA (most under driven), so about 34W per set.
See drawing.
If it is not enough, I might add some warm whites. See dotted circles. But no plan to yet.

[inkidu]

Not that you need my approval but I think your are on to something. Hitting particular wavelengths

but still covering large sections through the use of multiple colors. Take it for what its worth I

still would suggest the warm whites. Call it covering your bases also there is a ? in my mind how much

different carotenoids play apart in the overall system.

"There are over 600 known carotenoids(thanks wikipedia)" I would assume that different kinds would

respond to different wavelengths.

What kind of drivers are you using? I ask cause I am dealing with this right now with the deep reds.

A driver with a dimming function would allow you to test if favoring one type of chlorophyll or the other

would be beneficial. The reefkeeperlite can be set up to adjust 0-10 vdc dimmers. no experience with

Just a suggestion. Hope this helps.

[rygh]

I have a meanwell LPC-35-700 that I bought a while back, so will probably use that.
However, it is really running the deep reds at half power. Doubles the cost/lumen.
If you want to drive fully, I recommend a meanwell LPC-60-1400. 60W, 1400mA

I am having second thoughts about the warm white as well.
And about driving at half rate, since it really adds to costs.

[SantaMonica]

A lot of mentions of power efficiency, but truthfully if I would instead maximize filtering performance per unit size. My acrylic unit, 25 X 7 X 6 is about 1050 cubic inches. The same amount of screen area and filtering power, if done properly with led's (or fiber, etc), could be placed in a 3 X 10 X 10 unit, which would be the size of a dictionary. Power supply would be external.

[rygh]

Just a note that various other LED threads are recently saying BLUE is the way to go.
One was after someone talked to someone who worked raising algae, then posted it on the internet.
So it must be right!

But it could be. Red wavelengths get filtered by seawater, blue bounces around a lot, so things adapt.
I have plenty of algae in my main tank, and it has very cold blue lighting.

Enclosed is the spectra from CFL/LED that I found online.
Note the serious lack of blue in the LED, and a lot in CFL. Especially phillips.
The LED was from some head-lamp, which most likely means cool-white. No mention of brand or anything.

I have a prism somewhere in the house. If I can find it, I will do some measurements myself.

[whites]

Hey Rygh what threads can you point me to them? on this site?

[rygh]

Yes, on this site.
Board index / Algae Scrubbers / Algae Scrubbers / LED lighting.
Active, so at top.
Various posts in there reference some interesting information outside.

[inkidu]

I got this from another site. I find it interesting and possible helpful ( like anything on the internet ? first)

"Plants only require red light for oxygen evolution (turning light into energy). Blue light is only required for responses that shape the plant's growth such as stem growth inhibition and leaf shape and orientation. Some plants have responses based on input from ranges outside red and blue (read: cucumbers).

The wavelengths most people show for chlorophyll are in vivo. The in vitro wavelengths are actually higher, around 650 nm for chl b and 670 nm for chl a. Furthermore, one study I read showed that the normal a:b ratio of 3:1 can be altered up to 6:1 with more desirable effects using that much more a-stimulating light. That being said, exclusive 660 nm light seems to me the best available solution at this time due to the unavailability of 650 nm or 670 nm LEDs.

The PAR standard calls for 25 moles/m2 per day. This converts to about 440 W/m2 (this is my own math based on this document), though I suppose the 400 W/m2 stated previously in this thread is also adequate. Optimally, you want your red light split 3:1 (300W:100W) for a standard ratio (or 6:1 - 343W:57W - for a possibly enhanced ratio) between 670 nm and 650nm, respectively.

Blue light responses are maximized with 8-20% of the red light (so 400 W/m2 red would require maximally 80 W/m2 blue). These responses have a broad range with multiple peaks between 430 nm and 480 nm which means pretty much any blue LED will do fine. (I'm aiming for 460 nm)."

Wikipedia says "Measurement of the absorption of light is complicated by the solvent used to extract it from plant material, which affects the values obtained"

Is there some etiquette for copying a post from another site? Posting links? I at least think I have not taken this to far out of context.

"Red around 650 and red around 670 are still both red and, in a living plant, are the peaks for stimulating chl a and chl b. The problem with using something lower than 650 nm (such as the 630 nm reds) is you are mostly only stimulating chl b. This causes an imbalance in the chl a:b ratio when compared to plants grown in sunlight. Also, because there is more chl b in PSII than PSI, the rate at which PSI can process electrons falls behind the rate at which PSII can supply them. What does PSII do with this energy that it can't give to PSI? It begins non-photochemical quenching in the xanthophyll cycle in which the carotenoids take the excess energy and dissipate it as heat. During this process one carotenoid (such as violaxanthin) is changed into another (such as zeaxanthin). Eventually you run out of carotenoids to convert and dissipate energy and once that mechanism is saturated the plant begins experiencing photoinhibition. The charge caused by the light exposure has no where to go so eventually causes irreparable damage to the D1 protein in PSII and the whole system stops working. Thus, you have a plant with unbalanced systems and a bunch of broken PSII floating around."

This person's comments do seem to address some of the problems I think people might be having.

Correct wavelengths(especially considering the narrow wavelengths of todays leds)

The right ratio.

By the way I got my (6) 10 watt deep reds running yesterday with the meanwell CLG-60-24 2s3p

I want to test the voltage across each branch, to make sure any one branch is not out of sync, but even with sunglasses they are to bright.

Hope this helps.

[rygh]

This would also explain why warm white LEDs, don't really work that well either.
Looking at that curve, they are FAR stronger on the shorter wavelengths.
- peak at 600.
- looks to be about 30% at 670.

It also means that adding warm whites to cover other wavelengths might in fact be counter productive.
Unbalances the red.

I might order some deep reds soon as well. But been busy on my sump.

For LED brightness issue:
I always do it this way:
1) Solder 4" colored red/black wires to the LED first.
2) Then mount to heat sink.
- once mounted, a pain to solder, and I have a very good soldering station.
3) Drill holes in heat sink, and run wires through hole to the back.
Then, you do all the wiring, testing and everything on the lead wires, on the back side. Never touching or near the LEDs again.
Solves the blinding light issue. Also really reduces chances of damaging the LEDs.
(Well, unless you stupidly set the led's face down on a screw lying on the workbench. GRR)
So really a 3.5) Put something on the LED side of the heatsink/etc to protect the LEDs.

BTW: Was there an ESD protection diode on the ones you bought?
- I did not see it on one I was looking at. Bothered me. Be careful of cheap star manufacturers.

For that matter, where did you buy them?
- Seems like a lot of places are out of stock.

Not sure the voltage across an LED will give you the exact answer though. Vf can vary a fair amount.
With what I think you are doing, often good to have a precise resistor in the branch, on each chain.
- gives you something to measure, and helps load-balance. But burns power.
But I never bother. I just do a single chain on a simple constant-current supply. Costs more, but lazy I guess.

[inkidu]

This would also explain why warm white LEDs, don't really work that well either.
Looking at that curve, they are FAR stronger on the shorter wavelengths.
- peak at 600.
- looks to be about 30% at 670.

It also means that adding warm whites to cover other wavelengths might in fact be counter productive.
Unbalances the red.

I might order some deep reds soon as well. But been busy on my sump.

For LED brightness issue:
I always do it this way:
1) Solder 4" colored red/black wires to the LED first.
2) Then mount to heat sink.
- once mounted, a pain to solder, and I have a very good soldering station.
3) Drill holes in heat sink, and run wires through hole to the back.
Then, you do all the wiring, testing and everything on the lead wires, on the back side. Never touching or near the LEDs again.
Solves the blinding light issue. Also really reduces chances of damaging the LEDs.
(Well, unless you stupidly set the led's face down on a screw lying on the workbench. GRR)
So really a 3.5) Put something on the LED side of the heatsink/etc to protect the LEDs.

BTW: Was there an ESD protection diode on the ones you bought?
- I did not see it on one I was looking at. Bothered me. Be careful of cheap star manufacturers.

For that matter, where did you buy them?
- Seems like a lot of places are out of stock.

Not sure the voltage across an LED will give you the exact answer though. Vf can vary a fair amount.
With what I think you are doing, often good to have a precise resistor in the branch, on each chain.
- gives you something to measure, and helps load-balance. But burns power.
But I never bother. I just do a single chain on a simple constant-current supply. Costs more, but lazy I guess.

Thanks for the info

Your point is well taken. I chose to match the (6)10 watt deep reds with only (6)5 watt warm whites

so as to not over do the red spectrum. Time will till if I should have tried something even smaller

( 3 watt ) The warm whites I am about to set up,waiting on nylon washers, are going to be run

with the eln-30 I could lower the current some. Will try to see what works.

(Correction) There IS esd protection. My fault they are so small I missed it at first. Mouser.

Thanks for help.

[whites]

Great Research!!! :D

[inkidu]

I found some, what I think is valid, info about the chlorophyll absorption spectrum.

------------------In organic solvents---in cells----------------------occurrence
(nm) (nm)

Chlorophyll a-------420,460--------435(670-680)-----------all photosynthesizing plants

chlorophyll b-------453,643---------480,650---------------higher plants and green algae

chlorophyll c-------445,625-------red band at 645---------diatoms and brown algae

chlorophyll d-------450,690-------red band at 740--------reported in some red algae



There is something also mentioned about a rare chlorophyll e that is found in

golden-yellow algae. I have notice some people posting about yellow algae in

this forum whether it is same ? or why it is even important just thought I' d mention it.

Hope this helps.

[whites]

Hi guys its been awhile since I was last on here, back on the mission agian Im pretty sure I can get what ever spec i need made in the quad band led just looking at some of the charts for the chlorophyll and what about lower than 400nm purple range Sm posted a graph a few pages back that had some high levels in that range.. any thoughts?

A light with a range of purple to high blue then orange to high red would this be the go?? maybe some white just in case??

Update on the 3w led globes I had, well they have been removed they may have worked if the spread was better they were just to pin pointed in the one spot at the height i could get them, having the effects of when a flood light is too close to the screen. just outside that radious was growing algea put the watt level imo was just not that great. since then I have jammed a single old t5 in there and am having beter results with algea on the screen and in the tank not really seeing a decline of algea in tank but a goni that has been difficult to open up has really started to extend alot now so im led to belive that water quality is getting better. ( sorry Im a fiend for not measuring water parameters)

looking forward to tweaking the the ats for better results..

Cheers guys for the path Im on...

[inkidu]

I just added 4 rebel royal blues(min 440 max 460 typ 447) and 2 rebel blues(min 460 max 490 typ 470) in the first row

of a 7 "x10.5"x3" inch al heatsink.(4 would suffice)The second row consists of (6) 10 watt deep reds (660) and

the third (6) 5 watt warm both ledengin. I am getting good growth, been only 3 days since the new addition,

but I am playing around with different forms of diffission. Right now I am using

decorative/privacy glass and ?wax paper?. My second al heatsink consists of (5) rebel

cool white and a red with a 10 watt red an a 10 watt warm white from ebay also

getting good growth. I have not used another type of filter or light for several weeks that

includes sponge or overhang, which I was using for mechanical filtration i.e. replaced

filter floss often. Still thinking about going vertical because even though my discus

seem happy I am wanting to see the algae scrubbing, mass growth, that I see from more

traditional vertical scrubbers w/ more traditional lighting. Like the idea of not being

concern with the amount of food I feed the fish. A couple tips take it for what it is worth.

I used screws and nylon washers to attach my leds because it is the best way to get rid of the

heat. But the adhesive pads seem like,never used, they would be much easier to diy and the

difference in heat transfer probably does not make that much difference. My thinking

was if you setup something that lasts as long as some of these leds, just do it. Still wish I did it

with adhesive pads. This leads me to my second tip none of this is cheap after you buy all the

hardware you could save some money and aggravation by getting adhesive pads.

I am in for say $500 for the lighting. More than I want to think about. My only solace is that I

will not be needing to replace these lights for many many years.

Hope this helps.

[rygh]

Regarding adhesive tape:
There are really 2 issues. First is a bit lower heat transfer as mentioned.
Second is that if they get really hot, adhesive can melt and let go. Since this are mounted upside-down, they will fall off.
I would say that over 5W, you really need to look into the data sheets and details. Probably fine if less.
It worked well on the heat sinks on the rams on my graphics card, for years.

Agree that screws are a DIY pain. And if the washer cracks/melts, you can get shorts.

I really like arctic silver thermal epoxy. Great heat transfer. Very permanent.
Easy to mix, loosely clamp with clothespins/tape.
As a bonus, it is electrically resistive, so even less chance of shorts.
But you do want to wait overnight for it to harden.
And to replace it, you have to literally grind it off. (Or simply put a new one next to it)

[rygh]

For thermal tape, sounds like there is even a pre cut version.
http://www.luxeonstar.com/precut-therma ... -p-457.php (http://www.luxeonstar.com/precut-thermal-adhesive-tape-for-luxeon-stars-12-p-457.php)

[inkidu]

The washers might split? Now I am second guessing the

cheap thin washers, which I used for only 5% of my project. Thanks for the heads up

I mostly used the nicer ones,thicker, from Fastenal. They made it a several week project getting nylon washers.

It was always a wait to get them. How hard is it to cut out a washer?

At one point I got fed up and bought the washers from the local hardware store washers.

The local hardware store ones were much thinner. Nothing I can't correct.

Now I am happy that I used a low clutch setting on my drill.

[rygh]

I have certainly cracked cheap plastic washers on other projects.
They might melt if you do excess soldering while installed as well.
But good nylon ones should not be a issue.

But I use epoxy for LEDs, so no direct experience.
No idea how hard cutting your own washers would be.

I use epoxy (no silver) to build my sump as well, and my boat, and so on.
So a bit biased.

[rygh]

I am in for say $500 for the lighting.

Ouch, but such is the cost of R&D.

But a cost comparison, given all the new data, is probably a good idea.
Lets take the "gold standard" system to be SM's great pre built one.
- Standard 96W T5 setup.
Lets make the controversial assumption that LEDs will work, and be 1/2 the power used.
And lets estimate LED emitters alone are $150, and last 7 years (50,000 hours)
Power:
- Assume lights on 18 hrs/day. Assume $0.20/KWH. (cheap CA)
- Standard system = 96Watts * 18Hrs * 365 days / 1000 = 630 KWH/year = $126 per year
- LED is simply half that, at $63 per year.
Bulbs:
- Standard = Assume 4 new bulbs every 3 months. = 4 * $8 * (12/3) = $128 per year
- LED emitters = $150 / 7 = $22 per year
Ballast/Driver
- Seems roughly the same, and lasts about as long. Lets not bother.
DIY/Purchased
- Buying a T5 light and reflectors is not that expensive. Hard to estimate DIY cost. Again, best not to bother.

So LEDs save (126-63) + (128-22) = $169 per year.

[inkidu]

Thanks for the reply

I make to many mistakes to use epoxy. Duh disease

"How hard is it to cut out a nylon washer?"

You missed my sarcasm.

Thanks for any help.

[inkidu]

Trying not to think about the money spent but your post about how this might be saving me

money is encouraging. Maybe some more detailed #'s

(12) rebels and (6) 5 watt ledengin roughly $7 apiece =-----$126

(6) 10 watt deep reds ledengin $30 apiece = -----------------$180

(2) 10 watt ebay specials about -------------------------------- $35 total

(3) meanwell 18 watt drivers $14 apiece-----------------------$42

(1) meanwell eln 30 watt driver---------------------------------- $24

(1) meanwell clg 60 watt driver----------------------------------$50

(2) 10.5"x 7"x3" heatsinks $15 apiece shipped-------------------$30

(2) pc fans I had but lets just say $10 apiece-------------------$20

screws+washers+tap bits------------------------------------------?$40?

grand total $547

As far as power goes my digital aquatics controller says I am pulling between .9 and 1.o amps with my lights only.

Potential total of drivers 144 watts.

I feel confident that I am getting some serious cooling given the size of my actively cooled heatsinks so

if the data sheets are right the ledengin have a 90% radiant flux maintenance at 100,000 hours with the current I am running.

The rebels are not to far behind at 80% at 100,000 hours. Wow That is so over the top I feel like I will believe it when I see it.

edited the rebels one spec gives 70% at 50,000 hours but that is at a high temp. I think I will get more because I cool them well.

Still quite a long time

ebay specials, only 2, probably junk but they do look bright. live and learn

As far as this working after several weeks with only a power head running, which feeds the scrubber, and some

air pumps there is absolutely nothing that is filtering,besides the ats, my tank of 5 say 5" discus and some cardinal tetra.

My fish seem content(in fact I brought them back from the brink of death), I feed pretty heavily, the pea green water disappeared,

and I have no algae in my tank. Thanks for all the help guys, I am going to call this a success. Wish my camera wasn't broken.

There is still a lot to do just to make things look better i.e. background, tidy up wires and setup, add more plants etc...

Tank is still fairly new.

Hope this helps.

[SantaMonica]

Yes pics, after you clean up. Especially the screen before and after cleaning.

[rygh]

Yes pictures please.
Quite annoying that someone high tech enough for that many LEDs does not have a digital camera. :D

Well, just got off Newark Site. Ordered 8 deep red and 4 blue Ledengin 5W LEDs. $127.
I am going with just those, no whites, running low power at 700 mA, so 25W, for half the system, so 100 sq inches of surface area.
Running low power allows me more LEDs for better spread, and they should last basically forever. But costs more.
I will have 50W of CFL-floods on the other half.
But I am going to mount each one on an individual heat sink (usual cheap aluminum L/U channel), so I can reconfigure it easily.
All that on a moderately loaded 65G tank. I don't expect mass growth. But once I convince the wife on a bigger tank....

Putting it on my brand new Rev-2 ATS that I got running this weekend. (I need to update my other thread on that)

Guess what, LedEngin now has "far red", at 740nm.
But luminance seems to drop off there, so did not make sense to me.

Interestingly, the aspect ratio, area, and power, is just about right for the SM-100. If only it works.

[inkidu]

If not the white consider a little bit of red-orange or a regular red for some balance.

And a rebel royal blue with the peak of 447nm fits in nicely right where your ledengin blue falls off.

Still some ? in my mind about the needs of the different carotenoids.

There are some significant carotenoids that peak at the low 600 and the low blue.

Take a look at page 6 (book pages 106-107)

http://www.life.illinois.edu/govindjee/ ... apter9.pdf (http://www.life.illinois.edu/govindjee/photosynBook/Chapter9.pdf)

Notice they mention the difficulty of getting in vivo #'s for the carotenoids.

Still thinking I could of used some reds and red-orange with the deep reds. The most common

absorption charts in my opinion over simplify the perception of what the plant needs, wants, or can use.

Just suggestions.

Hope this helps.

[Amphiprion]

Of course they oversimplify it. They mostly focus on peak spectral absorption. If you'll look at absorbance spectra graphs, you'll see it is more of a continuum than anything, with the lowest being in the greens. However, all photosynthetic organisms are capable of using other spectra and have various accessory pigments (like carotenoids, but many others as well). These can absorb other spectra and successfully transfer electrons, therefore being quite useful for photosynthesis. That is why I usually recommend full spectrum (or close to it). Yellowish-white light is most similar to full-spectrum sunlight and usually has the most PAR. It may not be the most efficient way to provide the the necessary spectra, but it is the most effective.

[SantaMonica]

For a really useful DIY, how about someone build a test of ten different led's next to each other, all the same except for spectrum, and see which one grows the most. Then use the most-growing one in your scrubber.

[rygh]

Yep, thought about that.
After all, I currently have CFL and LED next to each other as of last night, so stay tuned for results there.

The problem is, it is not that simple.
- It is not exactly a cheap thing to do.
- LEDs of different spectrums have different luminance.
- You also need to test combinations of wavelengths, not just one.
- No single vendor has all the wavelengths to test.

But I may think about it some more.
One advantage to my setup is that it is 35" long, and only 6.5" wide, so pretty easy to test different things on different sections.
And if I mount each LED to an individual heat sink and mount, it might be easy to move things around.

[rygh]

If not the white consider a little bit of red-orange or a regular red for some balance.
And a rebel royal blue with the peak of 447nm fits in nicely right where your ledengin blue falls off.

I really want to try the "simple" version of basic deep-red + blue first.
And deep-red I think more of as an optimization, not truly different from regular red. So no real need for other reds.
The lumens on royal blue really drop off, so not very efficient.

[inkidu]

lm is what the eye perceives. Radiometric power (mW) is something different. From what I have read so far they are hard to compare.

Thanks for the reply.

The cool whites I use do work. I do believe you could consider them to some extent to be yellowish-white.

Let me try to be as clear as possible by what I see is the problem by oversimplify what is very complex.

If you put 8000 lumens,of white light, one inch away from a screen and there are nutrients available, you will grow algae.

That is simple enough.

If you use colored led lighting, especially when you consider the very narrow frequency of very expensive

color leds, and efficiency is of importance than there is a need to be as accurate as possible with the selection of the leds.

This in my mind is where there is some grey area. I do not believe when you consider the very specialized nature of

color leds and when you considering all the variables( sw or fw, flow rate, vertical or horizontal, nature of your base water, the

variability between leds) that a choice of what type, color, and ratio can be easily made by the most common graphs.

There is also the ? of the how the screen will mature though different types of algae.

Lumen with leds cost a lot of money. If you wanted to be as efficient as possible than color leds seem like what you would want to use.

I am using a power head that is pushing 475 gph at 0ft.

I have a couple small whisper air pumps. I can't be using many watts. If I was not heating my tank to 85 degrees

This setup would be next to nothing to run. I am only pulling 1 amp out of the wall to power my lights.

Hope this helps

Santa Monica,

Maybe an assortment of very specific/narrow color temp (ANSI) leds that luxeon sells. I realize you do not favor a horizontal scrubber but that

setup might be useful for a test i.e. line them up along the length.

Thanks for any help.

[SantaMonica]

Yes if you are just test spot-growth from each led, then it does not matter the layout, so long as the flow is the same for all.

[rygh]

To clarify the whole lumen/mW notion:
Lumens = mW / 683 * "luminous efficiency of humans".
The luminous efficiency peaks at 555 nm (=1), and falls off to 0.0001 at 380nm / 750nm
You need to ignore the "luminous efficiency of humans" in these calculations, since of course, this is algae we are talking about.

So royal-blue, even ignoring the efficiency, still drops off badly.

Agree: The whole point of the careful spectral research here is to optimize power efficiency.
- And at $0.25/KWH these days for my bills, that quickly turns into real money.

That is another of the key reasons I will run at 700mA. I could go to 1400mA, but I do not get nearly a 2X increase in lumens.
I pay for that up front, by having to buy more LEDs, but I save in the long run.

[SantaMonica]

I think indeed that a working version will need many low-power emitters. A continuous even spread of light is critical; even on my 100, if you look at the ends, growth goes from green to brown in just a half inch, because of the drop in light at the end of the T5.

[rygh]

Well, LEDs have a surprisingly good distribution pattern.
Cree has about a 20% difference across 80 degrees.
LedEngin is a bit worse, with a 20% difference across 60 degrees.
Thus an LED that is put 4 inches from the screen can quite nicely cover a circle 6.5" in diameter.
So I you distribute them 6.5" apart, you actually get double coverage in between.

From my first test setup, I did not see any difference in growth across the screen.
On my Rev2 build, I added a micro-diamond lighting diffusion panel for the top, instead of simple clear acrylic.
It definitely gets the light bouncing around in there. But I do lose some light.
Plus, everything is painted white. And flat white paint actually bounces light better than most reflectors. It just bounces it in a random direction.

Reminder from earlier in the thread: I measured the light distribution from a CFL flood lamp, it it was not particularly good either.

So certainly an issue that cannot be ignored. But with a little planning, quite solvable.
And I think the low efficiency and DIY pain of the small emitters make them a non-starter.

[inkidu]

How about this to diffuse the led light.

http://www.guardian.com/en/na/PatternData.pdf

I mention it because I want to mix the different colors as

much as possible while putting them as close as I can to my screen. Still thinking about going vertical.

Good transmission with decent diffusion?

The cost might be high but, if I can put a system together that runs efficiently and has less/easier maintainess,

I will live with the upfront cost.

Any opinions? (I want to stay away from acrylic for light transmission)

Glass is easier to clean and want scratch as much.

Also looking at the Whiteoptics/White97 for refection and dur-iflex as possible cheaper alternative

to the glass for diffusion as well.

Big ? is the cost.

Any help would be appreciated.

[rygh]

That diffusion plating looks interesting. Did you find a place that sells just a few square feet to a hobbyist?

I tried the small micro-diamond diffusion grating.
Basically, the cheap stuff with small diamond shaped bumps that you get in the lighting department of home depot.
I am not very happy with it.
It spreads the light around great. Works well in that aspect.
But it reflects light back like crazy. Pretty bad for transmission.
So I am rethinking that.

Not sure why you don't like acrylic.
Generally transmits 90% for a 1/8 piece. Same as 3mm glass.
And for a simple ATS shield, you can easily use 1/16 or even less. (Home depot picture frame acrylic. Super cheap)
There is a big loss in the UV range, but who cares.

My latest plan is just to move the LEDs back a bit from the screen, and go back to simple very thin acrylic.
And make sure everything, especially sides, is painted white inside.

Interesting, another new problem I have is condensation. Those water droplets get pretty big on a semi-horizontal system.
And without the heat from a normal light, don't really evaporate.
It was not really an issue on my old vertical. Another one of those tradeoffs to remember.
Not sure how to solve that. I really want a sealed system.

[DEEPOCEAN]

Hi guys

There is a LED lamp that has been compared to a 150 w metalhalide on severeral tests. Im going to buy santa monicas 100 ATS and was considering using these LEDs what do you think. 4 on each ATS should do the work or perhaps a L60 high power if it comes in a freshwater version. The freshwater version gives 8000 kelvin. could this work?

http://www.fish-street.com/weipro_led_lighting_led-c
http://www.fish-street.com/fish_street_ ... g_in_china (http://www.fish-street.com/fish_street_release_the_first_led_lighting_in_chin a)

Moby

[inkidu]

Led lights that are best for corals and appearance are usually cool white and blue/royal blue while

great for that purpose maybe are not so efficient for growing algae compared to a purpose built led grow light.

While it doesn't look like a bad product, I noticed that the type of led is not mentioned, which could make a big difference.

Also because leds are so directional some light measurements can sometimes be misleading.

While maybe not false a comparison is difficult if you consider that some lights output are more uniform over a larger area while

led if not blended well can vary quite a bit over a given smaller area. The measurement given is only what falls on that tiny sensor.

Most of what I have done so far has been diy leds ( best results from the deep red (660nm), white and blue combo) which could be made to

work with SM scrubber but this would entail diy work.

You would give up the nice sleek design of SM scrubber with some of the aftermarket led grow lights i.e. the kind that

screw into a socket. They are usually those underpowered and inefficient 5mm leds but there are some better ones coming out now.

I will give it some thought and I might be able to come up with a recommendation.

Hope this helps.

[DEEPOCEAN]

Thanks appreciate it. If there is any solution to go low wattage on Santas scrubber then im game.

[rygh]

Agree with the concerns.
Even the freshwater 8K may have too much blue, and not enough red.

The LED efficiency is a big concern as well. While they do not show the exact LED, they do have the count and picture.
Unless I am misreading the questionable english, the L60 version has 120 white , and 60 blue, at 21 Watts.
That is 117 mA per LED. They are clearly 5 mm high power LEDs from that, and the picture.
The problem : Those 5 mm LEDs are half the efficiency or less.

Plus, most of those are 100mA max. So that calculation on 117 mA is a bit scary. Might be overdriven.
Something that saves cost, increases brightness, but seriously affects long term reliability.
Combine that with the bad english....

SUGGESTION:
First, get the SM scrubber going with standard T5 bulbs as recommended.
Two advantages:
- It proves it works and so on.
- It allows more time for new products, or maybe a DIY build.

[inkidu]

First off none of this is close to being cheap and I have not explored ugly, for this application, screw in led grow lights,

but you asked. I am far from an expert these are just suggestions.

Also take into consideration that before these leds are barely into there prime years there will be something better.

Very rough draft for one side and please suggestions ( I believe one weakness is spreading out the light a square screen would be easier to setup)

(16) 5 watt deep red ledengin at $11 apiece (shown as a 0) http://www.ledengin.com/products/5wLZ/LZ1-00R205.pdf ($180)

(7) cree xpg at $8 apiece (shown as a " ) http://www.cree.com/products/pdf/XLampXP-G.pdf ($60)

(1) meanwell cen-100-54 http://www.meanwell.com/search/cen-100/default.htm ($60) (very new but available) 91% eff.

(1) meanwell lplc-18-700 http://www.meanwell.com/search/lplc-18/default.htm ($15)

(1) 24" heatsink (might need to shave down some width) http://www.heatsinkusa.com/storename/he ... 910-3.aspx (http://www.heatsinkusa.com/storename/heatsinkusa/ViewDept-263910-3.aspx) ($60) shipped

Add say $20 for thermal adhesive pads,wire,some quality diffusion(might be very important for this application and

also might let you use less leds) etc....

Thats in this configuration ......................00"00"00"00"00"00"00"00..... with the xpg slightly above keeping the deep reds spaced all the same

I thought of using the cree xpg because they have a lot of 450 nm blue plus there over the top efficient.

A very approx. total $395 for one side. Plus all the diy ( still much less work than if you would do small leds i.e. 5mm)

WOW

You can make the decision if that even approaches what makes sense.

One problem that I think happens is that people see these kind of lights on ebay and believe there trump up #'s and think they will

work but when it comes to people who have success you must realize that you must go big or it probably won't work.

As far as I know the ledengin deep reds are the only high power deep red leds. Everything else,as far as 660nm, are cheap low amp leds.

One weakness is that is a lot of watts over what it replaces although you might be able to run it less.

A square screen would be easier to diy as far as having a good spread of light.

Hope this helps.

[rygh]

I am confused by the "lot of watts over what it replaces"
My experience so far seems to be that LEDs will use about 1/4 the power of a CFL flood, so maybe 1/2 the power of a T5.
Obviously very rough guesses, but nothing to indicate I would need more power.

What is the size of your screen, and how far away will you be putting those LEDs?

My original 15W version seemed about right for a weak-flow, single sided, 50 sq inch screen, about 4" away.
Grew a bit better than the original 2 x 25W CFL flood lamps. But those had to be moved well back due to weak flow, so not the best comparison.

My current design in progress is a high performance, 50W LED setup, 40 GPH/inch flow, single sided 200 sq inches, about 5" away, optimized spectrum.

[inkidu]

My fault rygh I have been taking this some distance from your original post.

He mentioned using leds for a SM scrubber.

Using your thread to talk about my thoughts on leds.

Your setup sounds good waiting to here some results.

Might have mentioned but what drivers are you using for your new leds?

Thanks for any help.

[rygh]

I think diverging from the original post is great. The whole thing is a work in progress.
My new design is changing as well. Maybe this weekend I will write up some detailed stats and pictures.

My driver is a Mean Well LPC-35-700W constant current driver.
So each set = 8 deep red, 4 blue, plus driver, plus some cheap purchased heat sinks = about $150.
An SM-100 would only need 1 set. (50 inches, dual sided)
For my 200 sq in one, it would be $300.

I am also changing my hours from 18/6 to 8/4/8/4. Even the current 15W system seems to be burning the algae a bit.

[SantaMonica]

Let's make this the definition of a "successful" LED scrubber, or any scrubber:

1. Is the only filter. This includes no waterchanges, no vodka/sugar, etc.

2. Keeps N and P unmeasureable continuously, with regular feedings, for months at a time.


This will give a cushion when others try it, and thus it still will be strong enough to work for them. By "others", I mean people who:

Decide that half-size should work just as good as full size.
Let it go a month with no cleaning.
Think that keeping the light 12" away is fine.
Use glass as a screen.
And the one that I see every day: Leave the light on 24 hours.

etc.


If a scrubber cannot keep unmeasureable N and P by itself, and thus it requires other filters or waterchanges for help, then the scrubber is just not strong enough for what it is trying to do.

[rygh]

> Is the only filter.
Well, everyone has a biological filter already. Even a bare tank, but especially the usual reef setup with tons of live rock.
Plus, I feel it is good to run a protein skimmer automatically an hour/two per day, for safety.

> This includes no waterchanges, no vodka/sugar, etc.
No waterchanges is a really bad idea. Getting all the trace dosing just right is very difficult.
This will result in people doing ok for a while, then having various corals and such get oddly sick as various
chemical concentrations get out of whack.
But water changes of 10% per month or so, not 20% per week like some.
Agree with the no vodka/sugar.

> Keeps N and P unmeasureable continuously, with regular feedings, for months at a time.
Agree.

To be precise, maybe best to rank it in 3 categories.
A) Working:
* You get enough green algae growth on your screens to be worth scraping off every week.
Basically, this means the light/flow/screen is good enough for basic functionality.

B) Success:
* N and P unmeasurable.
* No use of chemical filtration, such as carbon, phosban, vodka, sugar, ozone.
* No need for minimized feeding and heavy protein skimming for nutrient reduction.
At this point, you have a nice healthy tank, which is what your really care about.
Fish/Coral should be healthy, well fed, and not chemically stressed.

C) Great success:
* Elimination of all unnatural filtration.
* Limited simple water changes.
* Algae growth in main tank gone, or receding.
Here, you are getting into aesthetics and levels of reduction in maintenance.
Great, but not really required for a healthy tank.

Even my small LED test system brought me to B and most of C.
It also brought me to your idea of success as well, since I did try turning skimmer off, and it was no problem.
But while algae growth has really slowed in the main tank, it is still there, so I
need to make my ATS larger and tune it to really out-compete, before I can consider it a great success.

[SantaMonica]

especially the usual reef setup with tons of live rock.

I should have said "except for rock or sand".


run a protein skimmer automatically an hour/two per day, for safety.

This does not do anything "for safety". It just removes 1/24 of the food of the day.


This will result in people doing ok for a while, then having various corals and such get oddly sick as various chemical concentrations get out of whack.

Not true at all. Algea is the natural filter and feeder of the entire ocean. The reason that the balances are what they are in the ocean is because the algae made it that way. So when you use algae, the levels get to where they need to be. Plus, many tanks, including mine, have gone a long time with algae-only filtering. Some 10 years. Matter of fact, it is the NON-use of algae that gets things "out of whack", because then you don't have the natural feeder and filter that operates the ocean.


You get enough green algae growth on your screens to be worth scraping off every week. Basically, this means the light/flow/screen is good enough for basic functionality.

This doesn't mean anything. If people are going to come to this site and build something that is "supposed to work on their tank", then it's irrelavent how much algae grows on thier screens. All that matters is how much filtering they get.


But while algae growth has really slowed in the main tank, it is still there, so I
need to make my ATS larger and tune it to really out-compete, before I can consider it a great success.

Correct.

[rygh]

run a protein skimmer automatically an hour/two per day, for safety.
This does not do anything "for safety". It just removes 1/24 of the food of the day.
When something dies and disintegrates, a protein skimmer can deal with that sudden surge, but an ATS can't grow that fast.
A short spike of N/P is no big deal, but if that kills something else, you can get a chain reaction.
But a skimmer only on a few hours a day may not do that much anyway. So I may have to rethink that.




This will result in people doing ok for a while, then having various corals and such get oddly sick as various chemical concentrations get out of whack.
Not true at all. Algea is the natural filter and feeder of the entire ocean. The reason that the balances are what they are in the ocean is because the algae made it that way. So when you use algae, the levels get to where they need to be. Plus, many tanks, including mine, have gone a long time with algae-only filtering. Some 10 years. Matter of fact, it is the NON-use of algae that gets things "out of whack", because then you don't have the natural feeder and filter that operates the ocean.
You might have missed the sentence just before that on trace dosing.
To clarify: I feel that water changes are valuable for the ADDITION of nutrients that get used by everything, including algae.
Sure, algae takes things out. But what puts things in? Food and air is about it. Well, and the obvious calcium/alkalinity.
For example, my iodine absolutely plummets without water changes / dosing.
Seems a bit worse after the ATS was put on, but one of my soft corals is growing like crazy now so who knows.
Then there are the others, like Strontium, etc. I don't want to deal with testing/adding all that.
Water changes are a pretty simple way of keeping all that straight.

Running a tank for 10 years with no other filter makes perfect sense.
But was that a reef tank, with no water changes, and only simple alk/calc dosing?
If so, it would sure save some hassles.

[SantaMonica]

When something dies and disintegrates, a protein skimmer can deal with that sudden surge, but an ATS can't grow that fast.

When something dies, it makes ammonia. That is what kills things. Skimmers don't remove any ammonia at all. Not even a little. Ammonia is algae's favorite food.


You might have missed the sentence just before that on trace dosing.

Sure, algae takes things out. But what puts things in? Food and air is about it.

I saw it. Everything you need to put in your tank is contained in the food you feed, and in the organics produced as the algae grows. Algae is 90% of all life in the ocean; it creates 100% of the food in the ocean, from the sun. Every trace element (which matters) that is in the ocean is put there by algae. That's why algae handles metals so well.


Then there are the others, like Strontium, etc. I don't want to deal with testing/adding all that.


my iodine absolutely plummets without water changes / dosing.

Try feeding more. If you fed anywhere near natural reef levels, you'd have plenty of iodine and everthing else.

Cal/Alk/Str are not trace elements. They are basics, like salt. And dripping kalk, or dosing ca/alk, is a lot easier than water changes.


Running a tank for 10 years with no other filter makes perfect sense. But was that a reef tank, with no water changes, and only simple alk/calc dosing?

Yes, everyone knows about the IA tanks.

[rygh]

Interesting data. I may need to rethink my "safety" features. Hmm.
No idea what an IA tank is, but I will search.

At any rate, I promised to post some new results and info. Following.

[rygh]

This is the new setup. See other thread Rev2 horizontal build for details.
I want to emphasize by re-posting this here that the left part is standard CFL flood, 2 x 24W.
The right is the original 15W LED system in the beginning of the post.

[rygh]

Here are the screens, with 1 week of growth. Only up for 2 weeks, so still getting established.
While it looks yellow, that is fairly thick yellow, with some ice-plant like pale green growth, not
a simple thin dusting.

It should also be emphasized that this 200 sq inches, on a 65 gallon lightly fed, protein skimmer on, system.
So the lack of huge clumps is completely expected.

Top left screen is the CFL section.
Top right screen is the LED section.
Bottom screen is a mixture.

The LED system clearly has an issue with light distribution, but it is really too small for this size screen,
so that is a fix in progress.

The LED side is a bit darker green, but I would say there is slightly more algae mass on the CFL side. Maybe 20% or so.
Of course, that is 50W versus 15W.

[rygh]

Here is the scrapings from the screen. Pretty good.
I also included the amount of food for the same period.
I am frankly pretty amazed by how close in mass they are.
Although the basic physics say they should be.
Still, it is really good to see.

[rygh]

ON TO THE NEXT REVISION!

As stated up a ways in the thread, with the larger system, I need more LED power.
So I am building a 12 LED, 25W, spectrum optimized system. --- hopefully.
8 x special deep red
4 x blue.
Running at 700 mA
So there will be 4 sets. Each set has 2 red, 1 blue.
Each set is on a single large heatsink.
I decided not to make my own this time. Lazy, and found some surplus. Hence the odd shape.

Here is the basic set.

[rygh]

Here is the box that holds the 4 sets, and the 4 sets basically sitting on it.
Very simply plywood box, painted white, holes cut where LEDs pass through.

[SantaMonica]

The reason you have yellow on the CFL side is because it is flow-limited. When scrubbers have enough light to grow, but not enough flow to transfer metabolites, it grows yellow. This is why the LED side grew green; the light-power is lower, so it has enough flow to handle it. Thus, the LED side is currently at its max filtering, whereas the CFL side is ready to jump up in filtering as soon as flow is increased.

[rygh]

Well, it is 6.5" wide, at 300 GPH, so 46 GPH/inch.
Pretty significant flow for single sided.
Measured with buckets/stopwatch, not a guess on the pump specs.

But I have a spare 250 GPH pump, and can add that temporarily, to see what happens. Easy enough to test.

Note:
That is not a thin bit of weak yellow. It is thick, with pale green lumpy stuff.
N and P are undetectable.

[SantaMonica]

Then pull back the lights, or run them two hours less a day.

[rygh]

Finally, my REV2 LED setup is done.
These pictures show the final test, and the box in place and running.

Side note: Lights are now on 7/5/7/5.

[rygh]

I am attempting to show the color and brightness, but the camera really has trouble with it.
The pictures show it roughly, but definitely do not do it justice.

The color is this really amazingly cool looking redish purple.
The deep red + blue is a really interesting combination.
Even if this fails to work, it has some interesting wow factor just on the color.

One of the previous pictures, of the LED system in place, captured the color well on a few points where it leaked out.

BTW: My garage is very brightly lit from some big overhead shop lights, so that little 25W system really puts out the light.

[whites]

you got the aroura lights in your shed man

[ocean rock]

hi rygh
iam hoping thats going to work for you it looks like it might do the job fingers crossed ,

[Gigaah]

I was pointed this way from another site. Perhaps I can help debug your scrubber?
Without reading the entire 16 pages worth of post. Can you give me the colors(in Kelvin or nm) of your LEDs, the area over which they are spread, the number and lumens(if you have it), or model and watts if possible.

Judging by a few pictures I would say that you almost certainly have the wrong color going on. I know terrestrial plants use the red end of the spectrum, but did you know some marine corals have been shown to have their grown stopped by red? I don't know for sure how algae reacts but I do know for certain that skipping the red certainly will not hinder things. The whites should be about 6400k. 5500k isn't great, 7-10k might do but 6300-6500 is the sweet spot. Benefit AND problem with LED's is they are such narrow range that you do not have to be far off to miss the mark, but when your on you are ON.

CFLs are all over the place so you don't have to be all that close to get SOME useful light out of them.

[inkidu]

My rebel cool whites(LXML-PWC1-0100) do pretty well especially considering that only one 18 watt

driver is powering them and I have so much more power on my other heatsink/setup.

And they have a typical 6500 k although that # varies quite a bit (don't know bin #)

I been eyeballing the new cree xp-g leds. I want to use some of my leds to begin lighting for plants.

The cree xp-g have, at least from cutter, known bin #'s and they just came out with neutral and warm white.

Around 250 lumen @ 700 mA for the R5 bin even if you lose efficiency by going with white instead of colored

that kind of output is hard to ignore.

(7) xp-g for $65.55(shipped) at Cutter doesn't sound so bad right now and I still have a 18 watt driver I am not using.

Hope this helps.

[rygh]

I was pointed this way from another site. Perhaps I can help debug your scrubber?
Without reading the entire 16 pages worth of post. Can you give me the colors(in Kelvin or nm) of your LEDs, the area over which they are spread, the number and lumens(if you have it), or model and watts if possible.

Judging by a few pictures I would say that you almost certainly have the wrong color going on. I know terrestrial plants use the red end of the spectrum, but did you know some marine corals have been shown to have their grown stopped by red? I don't know for sure how algae reacts but I do know for certain that skipping the red certainly will not hinder things. The whites should be about 6400k. 5500k isn't great, 7-10k might do but 6300-6500 is the sweet spot. Benefit AND problem with LED's is they are such narrow range that you do not have to be far off to miss the mark, but when your on you are ON.

CFLs are all over the place so you don't have to be all that close to get SOME useful light out of them.

Well, I think you really need to at least skim through the thread first. There is a ton of detailed data on what algae appear to like as far
as spectrum, and discussion on how best to meet that. It is quite different from coral.
Details of the LEDs/power are in there also. Although I may summarize again if I have some time.
The pictures really don't tell the story very much on color. Cameras have a terrible time with these odd colors.

[rygh]

Rev 2 LED details:
8 x Deep Red LedEngin. LZ1-00R205
- Running 700 mA (low power), 2.9Vf = 2W each
- About 100 lumens, but not very accurate for deep red.
- About 400mW radiant flux.
- Peak wavelength around 660 nm.
4 x Blue LedEngin. LZ1-00B205
- Running 700 mA (low power), 3.4Vf = 2.4W each
- About 50 lumens.
- Peak wavelength around 450 nm.

So 25W total.
About 1000 lumens total, but again, not a really accurate number to use with these colors.
Kelvin rating color temperature is not relevant at all.
This is currently spread over about 150 sq inches, although that may be increased to full 200 sq in.
No lenses, but there is a micro-diamond light diffusion grid. Standard lighting grid from Lowes.
It is running 7/5/7/5.

[rygh]

The big problem with white LEDs is that they are not white.
Not even particularly close.
There is a big spike in the blue at around 425 nm, and a broad spike around 560.
- Warm / Neutral / Cool are all a bit different, but still a similar concept.
Basically, they are actually a blue LED, which activates phosphors on the lens for the upper colors.

The problem is that the spike at 425 is a bit low for the blue aspect of chlorophyl.
And they drop off at 650 or so, which is a bit low for the red aspect of chlorophyl.

But the big problem with Red+Blue, as mentioned quite a few times in this thread, is that
since LED wavelengths are so tight, if you miss the key wavelengths, it may be completely useless.

My first Rev1 LED was a mixture. Warm white + red for the upper regions.
- That actually did fairly well. But it was simply too small for my tank.
For Rev-2, I am taking a shot at pure red + blue, with special deep-red, which hits chlorophyl better.
- Hopefully more optimal, but may end up worse. Results for the first few days are not fantastic.
There is a good chance I may need a rev-3. We shall see.
- One thing is that the deep reds and blues complement the warm white holes better than before.
- And I have warm whites from the first build of course.

[rygh]

Basic notes on algae and chlorophyl:
We are trying to grow green hair algae or similar such plant here.
Those plant types plant are generally seen in shallow water. Both fresh and salt.
As such, they expect fairly similar wavelengths as typical terrestrial plants.
Unfortunately, it is hard to say exactly what. Most of the spectral analysis that is seen is based
on what wavelengths the algae ABSORBS, which is not exactly the same as what it USES.

Corals, especially deeper water corals, are very different.
They have adapted to using much more of the blue end of the spectrum.
This is because most of the red end is gone by the time light gets to them.
Partly due to the weak absorption of red wavelengths by water itself, but more so by all the free floating algae.

So for the main tank, you want very little red, and lots of blue, to starve green algae, and grow coral.
For a turf scrubber, it is the opposite. We are deliberately growing the green, so want lots of red, and less blue.

[inkidu]

Rygh, I here what you are saying. My thinking is that CREE spends a lot more time

and resources in making white(blue w/ phosphors) leds than any other led. Hence they are over

the top bright/efficient. Red leds are sometimes a different type of die.

They don't even make a colored xp-g. I might be wrong

trying to figure it out myself. Understandably the whole thread has mostly been about

what will be the most efficent ???

Thanks for any help.

[rygh]

I agree that cree seems to be the leader in lighting FOR PEOPLE.
They do a great job at the most efficient lighting for our eyes.
And even the "luminance" measurement includes a factor for the spectrum used by human eyes.

But all that has little correlation with what is best for algae.
And I think that fact is why a lot of LED driven ATS setups fail.
You get fooled by what looks to us light great lighting. But it is terrible for algae.

On the other hand LedEngin actually mentions this in their deep-red datasheet:
"The LZ1-00R205 LED emitter provides superior radiometric power in the wavelength range specifically required for plants’ chlorophyll a absorption."

For LED lights I put in around the house, I have definitely used Cree.

All that said, I cannot definitively say that what I did was necessarily correct either.

[Gigaah]

I've spent a lot of time pouring over spectral graphs about lights and LED's, LED grow lights, etc.

I suggest 470nm blue and 630nm LED and/or 6500k and 4000k. I think your missing the spectrum marks just a little bit.
Other marks to possibly hit are 430 and 660(you already have the 660's covered pretty much)

Cree does make a range white emmiter(think its 3 or 4 emmiters in different K ranges) and also an RBGW that has 4k white that would probably be good. prices? Probably not cheap.

[rygh]

I am a bit worried about the blue as well. The red I think is ok.

There is an extra problem that should also be emphasizes: The accuracy is terrible.
When ordering just a couple, you get what you get. It is rare be able to choose detailed bin code.
You have no idea where they fall in the bin.
They vary a lot based on current and temperature.
For example, my BLUE is from a very good vendor, but still may from about 440 to 490.

I am going to wait a few weeks for results.
But if it is not great, then I may try a "shotgun" approach.
Basically, add a bit of everything and make a fairly flat spectrum.
One downside is wasted power, but it is still a lot more efficient than CFL.
Another downside is that the optics become a bit tricky. You cannot have a lot of each color, so even
distribution becomes a pain.

[inkidu]

I just ordered (8) xp-g warm white leds $63.73 (shipped) from cutter down under

XPGWHT-L1-7D2-Q5-0-01 Warm White 7D

Planning to combine them with some ledengin deep reds.

If they don't work so well I still have some of the best leds made right now.

If any one wants to do some R&D please do. This is getting expensive.

Will see if this helps.

[Gigaah]

Do you have a spectral graph on that one? I've seen warms with very tall "royal blue" spikes and ones with short "blue" spikes
"

[inkidu]

http://www.cree.com/products/pdf/XLampXP-G.pdf (page 3)

XPGWHT-L1-7D2-Q5-0-01 Warm White 7D

................7d2.........................

is the bin # that describes chromaticity

http://www.cree.com/products/pdf/XLampXP_B&L.pdf (page 11)

Hope this helps

[Gigaah]

Yeah thats one of the ones with the small blue spike and the really high sweeping red spectrum. If its red your after then that certainly is the one.

I've got an itch to wire up the red and the blue that I think would be best and try it out on my scrubber inplace of one of the CFLs for a test. Gunna have to check cost on how much that experiment would cost me.

[inkidu]

I am thinking this will be my next purchase ledengin deep reds and (10) of these 680nm leds for $90

to go along with the xp-g warm whites I got coming.

http://www.tech-led.com/data/SMB680-1100-I.pdf

Might be better choices through this company, but this is what is available in a min. 10 led sample.

Seems that some companies are only interested in large sales.

Will use a meanwell cen-100-48 to drive the 680's

http://www.meanwell.com/search/cen-100/default.htm

Still doing research but I think I will be able to mount them on a small metal "pedestal" with

Arctic Alumina Adhesive (non conductive) letting the leads "dangle"

I think the heatsink part of the led is non conductive. The I suffix indicates a ceramic plate between

the led and the heatsink. ??? anyone know for sure?

Needless to say this is getting farther and farther away from a basic diy setup.

So please any comments and/or recommendations.

One concern I am having is that rough calculation has me using a lot more power over a 660nm

Have 6 10watt (660) deep reds 833mA * approx 11v *6 leds= 55 watts giving approx total of 12 watts of radiant flux

Plans for 10 (680) deeper reds 2A * approx 4v *10 leds = 80 watts giving approx total of (420mw *10) 4.2watts of radiant intensity

(assuming that is the same as radiant flux) (edited) radiant intensity is mW/sr ??? I am guesstimating around 400 mW

Might go to show how good the ledengin deepreds are and that the further up the scale you go the more power you need to use.

Thanks for any replies.

[rygh]

All of the "star" type LED PCBs are non conducting. You will not have a problem with those. Not sure about others.
I use Arctic silver or Arctic alumina. Works great. (I use the epoxy version and glue it, but same as the grease)

I bought my deep reds from Newark. No real minimum order. Decent pricing. But no idea on bin codes.
I agree that they will complement the xp-g very well.
I have no solid opinion on the ratio. My first build used 3 warm-white to 2 red. Equal wattage.
Worked so-so.

You might want to double check the blue side as well.
Those xp-g have a fairly low and tight blue. You might need a royal-blue to be perfect.
Assuming you are going for a fairly flat spectrum.

Holy smoke, 135 Watts?
I forget if you mentioned it, but what is the area of your screen? It is starting to sound like a serious amount of light there.
I think my rev-2 at 25W might be capable of the entire 200 sq in screen. It seems to be overpowering the 120 sq in or so now.

[inkidu]

I hear you. 6" by 36" horizontal

Plan on moving some part of my led setup, they are on different heatsinks, to light my tank.

I think I might of overdone it a bit with the (6) 10 watt deep reds (6) 3 watt blues and (6) 5 watt

warm whites on a heat sink that measures 7" by 10.5". Although I kind of just thought I would just leave it on less.

To much work(backwords) to disassemble so I will just move it.

Want to get some plants going. FW discus south american biotope action.

Might be a good point for others we both allowed some measure of flexibility in our led builds for

adjustment by using multiple heatsinks.

Hope this helps.

[rygh]

Keeping the LEDs modular is a good suggestion.
I originally had 1 per heat sink, and on the latest one, 3 per heat sink.
Wish I had stayed with 1.

That will be a really funky looking purple main tank if you keep that 6/6/6 R/W/B color ratio.
Cool!

You have an adjustable driver. Just drop it down. More efficient watts/lum that way.

I am also going to use some leftover warm-whites for my kid's simple planted FW tank.
He rather likes the funky purple also though, so who knows.

[Gigaah]

Keep us results posted. I have some LED comming that I'm going to give a shot.

I'm just as curious about the far red spectrum as I am about the deep blue and uv spectrum(blue mainly in marine life).

"Chlorophyll a fluoresces at 673 nm (maximum) and 726 nm."

[inkidu]

Time to get funky on the dance floor.

[attachment=0:1zbgw4il]012.JPG[/attachment:1zbgw4il]

[inkidu]

This camera is causing me nothing but problems. With new batteries I still can only manage a few photos

before it dies but I managed to get a few more. The tank still needs a lot of work it is still pretty new.

This is about 3 days growth and is only one half of the full scrubber, the side with the deep reds.

[attachment=1:1a7wyp5p]017.JPG[/attachment:1a7wyp5p]

[SantaMonica]

That may work out.

[Gigaah]

Awsome. The last week I've also been doing some hard research and looking at deep red. Here is a link that seems to shed a possible why onto the red spectrum and possibly even cyan green yellow.

http://www.mbl.ku.dk/mkuhl/pages/PDF/La ... l_2005.pdf (http://www.mbl.ku.dk/mkuhl/pages/PDF/Larkum&Kuhl_2005.pdf)

My suspicion is our CFLs have been putting out enough of the spectrum we figured was kinda useless to grow the algae. where as our LEDs are just so narrow they do not. especially the specific color ones.

Here is another semi usefull titbit I found from the bio fuel algae guys.

http://www.oilgae.com/algae/cla/cla.html

It maybe be possible with enough work we can dial in on how to specificly promote growth of selected types of algae. as well as use even less wattage.

The work we are doing here also has implications outside algae. It might help discover more about lighting for coral. If new wavelength targets are identified in the green or deep red spectrum.

[Gigaah]

I will be targeting warm white roughly 3000k and 700nm for my experiements. I just have to wait for them to arrive.

[inkidu]

I found this interesting and a reason to go with the 680 nm led and 5 watt 660 nm deep red.

b) Emerson effect ( this is from here http://www.photosynthesisinfo.com/moder ... synthesis/ (http://www.photosynthesisinfo.com/modern-concept-for-mechanism-of-photosynthesis/) )

Emerson and Lewis (1943) measured quantum yield at different wavelengths of light. Quantum yield can be defined as number of O2 molecules released per quantum of light absorbed. A sudden drop in rate of photosynthesis was noticed at 680 mµ (red region). This sudden fall in the photosynthesis yield beyond red region of spectrum is called red drop.

Emerson et al. further noticed that photosynthetic rate can be restored if simultaneously shorter wavelength is provided. This simultaneous giving of shorter and longer wavelengths gave photosynthetic rate higher than total rate from the beams separately. This photosynthetic enhancement is referred as Emerson enhancement effect or Emerson effect. The results obtained by Emerson were as under:

(for some reason I can't upload graph but more or less 1+1= more than 2)

Photosynthesis at 700 nm = 10

Photosynthesis at 653 nm = 43.5

Photosynthesis at 653 + 700 nm = 72.5


With there combined spectral power distribution curves(660 and 680 nm) and setup to have a similar light output, I am

going to shoot for around 400 mW per led, I might be able to achieve a fairly consistent output between 660 and 680 and will

have about a .65 relative spectral power for 650 nm and a .2 relative spectral power for 700 nm.

Still some concerns about if the benefit is worth the 80 watts I will be using for the 680 nm led. (edited around 400mW ) 2A * 4V * 10 leds

In comparison the 660 nm ledengin would be: 1A * 2.8V * 10 leds = 28 watts for over 500 mW a piece

As always please any input would be appreciated I am far from an expert.

I will pickup some 650 from the xp-g warm whites also. I would hit the whole red spectrum from 600 to 700nm.

It would be nice to figure how much 650 is contributed by the xp-g's. I forgot ideally I want to stimulate the chlorophyll a (670) more than than the

chlorophyll b (650) and this would be easily enough done by dimming the xpg and/or increase power to the 680. I read that there is a way to calculate the contribution of the w/white

but it is time consuming. A person from another site explained it like this:"you would use the spectral output chart that shows the intensity as a function of wavelength and write down the relative intensity for each wavelength. Last time I did it, I used every 10 nm but the more sample wavelengths you include the more accurate your result will be. You would then have to set up an equation based on the total lumen output to discover the actuall percentage of lumens each wavelength contributes. Then convert the lumens of each wavelength to watts to remove the lumen weighting (which is why I used 10 nm increments). I suggest using a spreadsheet to do all the math for you."

[Gigaah]

I've read somewhat similar information. That in conjunction with the information I posted is why I decided to experiment with warm white and 700nm. The warms hit some blue, and sweeps over to get some yellow, lots of orange and red but drops off too much before 700nm which is why I wanted to put in a 700nm deep red.

[rygh]

Wow, Inkidu really has some good growth there. Major jealousy.
But looking at the irregular pattern, and the picture of the LEDs, you might have a light dispersion issue.

I am still just growing a pale yellow algae mixture.
It grows ok, given my low nutrients, but it is not that green.
I shaded a section of the screen with duct tape, and that section did turn green fairly quickly.
However, that section did not grow nearly as much algae mass.
I am going to try more flow, and maybe less flow, as well. Just to double check that it is not an issue.
I am also playing with on/off cycle a bit more.

It is fairly green under the CFL-flood test section as well. Although some yellow. But again, less algae mass there.
I tried the duct tape light blocker there as well, and it was more green, but started to die off as well.

Since this setup is with pure red+blue, and my original test system with red+white produced green, that there might
be a real problem with the missing spectrums.
It grows, but it grows the wrong type of algae. Not entirely sure I care, but I think I do.
I think if I did have a lot of nutrients that I would care, since I really doubt it grows as fast. Too solid.

N/P are still 0.
Algae in the main tank is still weakly growing. Not dying. Better than no ATS, by a lot, but not there yet.

[rygh]

I am really confused by the amount of power.
I have a 25W system which really appears to be too bright and frying the algae.
I have been busy adding duct tape to screen out regions
Yet Inkidu has a 135W system that seems to be growing great.
We both have almost exactly the same size screen.

Inkidu, some key questions:
1) What is your flow rate?
Maybe I am way off there.
2) Is it running full power?
You have a controllable supply. Just wondering how it was set. Measured?
Especially looking at that picture. My camera cannot handle anything close to a direct shot like that.

Maybe the power has to match the nutrients? That doesn't really make sense though.

[inkidu]

I have an Azoo Powerhead Model 1800 which supplies 475 gph, I would guestimate 380 with a 8" lift hard

to tell (powerheads don't pump up so well).

As far as I know you need to turn down(current) manual on any meanwell driver for it to be any different than what the

driver is spec at. So no I have not done any adjustments or for that matter measured i.e. I just planned my drivers to fit very well with the load.

That is also to say if anything goes wrong the driver would almost certainly turn off because it's out of the normal range. (just an assumption here)

Don't let the picture fool you they are really bright the picture is just being taken at a very oblique angle.

I hear what your saying about the uneven growth something I am playing with. I got a Lee sample pack of filters

http://www.leefilters.com/

I was not sure what to use so I just used an assortment of the diffusion film that came with the sample pack(only say 1.5" by 3" inch),

and that might be some of the reason that there is some splotchy growth i.e. different levels of diffusion. It is overpowered. I contacted the one

company about the pattern diffusion glass. Something that I feel will break led use for scrubbers wide open.

http://www.guardian.com/en/na/gp_016074.html You could use high power leds, therefore eliminating the diy headache, mix the colors well and

place the leds very close to the screen without the risk of hot spots.

I have been given the cooled shoulder by the company. I did find out that it is made in Spain and not in the USA (my home town)which is part of the problem

(they don't want to bother with small orders not that I blame them) but I am working on that. Using duct tape ??? If you were using typical lighting

I would go with what SM always tells people (hope you don't mine SM and I get it right) If you think there is to much light try to increase flow if you can,

if not move the lights back but since you are using leds there could be other factors involved. I wonder if a little more time for everything to find its place

might help? Trying to work it out myself.

Hope this helps.

[rygh]

Same flow as me. Thanks.
The side with the CFL flood is not doing great either, so I am really thinking that it is a simple matter of almost no nutrients.
I did turn the skimmer off and increased feeding a bit, but not a lot yet.

One thought I had on light diffusion is to light it indirectly. Have the LEDs on the side, pointed slightly up. Paint entire box flat white.
Flat white paint reflects great. Like 80%+, and in a very random direction. Super cheap and simple as well.
I did that on a regular light in my den. Replaced a halogen torchier light with an LED light, purely bouncing off the ceiling. Worked great.

I will keep tinkering. I might add the warm whites from my previous build.

Yes, duct tape. :lol:
Worlds fastest way to make a reflective surface.

[inkidu]

Still like this product for reflection. whiteopitcs film(can handle some moisture) every time I think of it its after 5pm

http://whiteoptics.com/?page_id=12 It might be expensive???

I am taking your advice and thinking about using more blue but want to expand my blue spectrum from what the 450nm blue the xp-g gives me.

I want something that hits one of chloro directly in this case b (still having some trouble finding what I want in a 430nm leds i.e. chloro a

"gallium nitride on a silicon carbide substrate" is what I think is needed for 430nm????) Still a ? in my mind about how much blue is needed.

- XPEBLU-L1-B40-K3-0-01 Blue 470-480nm 35.2 Lumen @350mA from cutter

[rygh]

There are plenty of 5mm round LEDs in the 420nm and even UV range, but those are weak.
Looks like Nichea makes some 2W version for 380nm, but I don't see a supplier.
Although a few for sale as flashlight bulbs.

Good paint won't really have a problem with moisture, as long as you spray it on acrylic.
I used some to paint PVC pipes dark blue in my main tank.
And for that matter, the paint can be on the outside.

[SantaMonica]

Trust me, you have enough nutrients, especially since the skimmer is not removing any. If you are growing pale green, then it has nutrients. Pale green is just burned, that's all. Probably because the 25 watts are focused into hot spots that's don't appear to your eyes. Just reduce the ON time to fix it.

The white reflective ideas has merit; is similar to the edge-lit ideas tossed around last year:

[rygh]

Trust me, you have enough nutrients, especially since the skimmer is not removing any. If you are growing pale green, then it has nutrients. Pale green is just burned, that's all. Probably because the 25 watts are focused into hot spots that's don't appear to your eyes. Just reduce the ON time to fix it.:
Sounds good.
Changing to 5/7/5/7 on/off/on/off, so 4 hours less per day.
Also, I am going to add a bunch more panels of diffusion grating.
Looking at my quick duct tape fix, it does seem to be growing best just under the edge, but not too far.

Longer term, I am thinking of a rev-3:
1) Extend LED the entire length. So no more CFL test area.
2) Mount LEDs on the side, angled up a bit, using white-paint diffusion idea.
3) Add 5 warm-white 3W LEDs.

[SantaMonica]

Looking at my quick duct tape fix, it does seem to be growing best just under the edge, but not too far.

This could also be from diffraction and mixing: as light hits a solid edge, it spreads out. It's the same pricipal as a pinhole camera, and might be mixing the reds and blues more.

[inkidu]

I managed to get (8) XPEBLU-L1-B40-K3-0-01 Blue 470-480nm 35.2 Lumen @350mA (cree xp-e)

placed on my order of (8) cree xp-g's before they shipped for $48.72 additional. Thats a $116.42 total

At least I didn't pay shipping twice.

I have a feeling these and the ledengin 660nm and the 680nm leds will make quite a setup. Those are going to be

about $220 plus around a $100 more for drivers etc...

Hope this helps anyone thinking about getting leds. Don't

At least when they become more affordable I will know what to get, what works, and how to use them.

I hope I am doing this right. Time will tell.

Side note where the xp-g drops off the most is where the new blue fills the gap i.e. 470-480nm.

Another improvement I have been throwing around is that the royal blue rebels are, as far as I've found yet, some

of the lowest (nm) and affordable readily available leds. The peaks of the rebel royal blue (447nm), the cree, the bin I bought, is (475) and

the cyan rebel peak is (505). So the combination/mixture between the three seems to me to cover a great part of the spectrum

needed for the low nm needs for the chlorophyll and good amount of the the cartonoids.

As always I am thinking aloud so that anyone could give me an opinion and/or corrections on my thoughts.

One problem I am seeing is that with so many blue leds, even at low 350 mA, I will need more reds.

Or another way of saying it more money and diffusion has always been something I have been struggling with.

Thanks for any help.

[rygh]

Looking at my quick duct tape fix, it does seem to be growing best just under the edge, but not too far.
This could also be from diffraction and mixing: as light hits a solid edge, it spreads out. It's the same pricipal as a pinhole camera, and might be mixing the reds and blues more.
Well, not exactly. Diffraction is only pronounced when the slot size is on the order of the wavelength.
But under the tape, it is purely reflected light, not direct. And it is reflected from the white paint, which scatters very well,
so yes, it is more mixed.
However, the red/blue/red LEDs are only a few inches apart, and about 6 inches up, plus I have a layer of diamond diffusion lighting grid,
plus, the growth on the screen is quite uniform.

So all evidence is that it really has to be mostly about intensity, not mixing.

Although there is a chance that the wavelengths do not reflect the same. Blue tends to scatter quite a bit more. (Blue sky)

[rygh]

One problem I am seeing is that with so many blue leds, even at low 350 mA, I will need more reds.

Maybe, but since algae seems to grow just fine in reef display tanks, which have a TON of blue and almost no red, I would
not worry. And in fact, it might be a good thing.
Although you may end up with some Coraline algae growth. ;)

[Gigaah]

I'll take a picture of the bottom of my scrubber tub(large walmart tote) its totally covered.

[rygh]

Regarding Inkidu's cost concerns.
I just put down $150 just for T5 bulbs for my main tank. And it is already 1/2 LED as well.
A decent actinic is $25, and only lasts about 6 months. (Maybe over driving was a mistake)
But yeah, sure seems that you like to spend money. 8-)

[inkidu]

Please don't remind me (cost) it is easier if I don't think about it. Cool picture.

Just now filling in the paper work to get the 680 nm leds.

I used a spectrometer to calculate/estimate the spectrum of ledengin deep red. Just curious.

Using a hydrogen lamp as a reference the deep red had a spectrum that was very defined on one side,

the side slightly higher than the hydrogen red spectra llne (656), and then gradually tapered toward the yellow end of the spectrum.

Which does match the spec in that the slope of the line seemed to taper off more gradually on the yellow side than the red.

Seems that most curves are balanced on each side.????

I also notice that the total/whole spectrum was much tighter for the warm white led in comparison to an incandescent bulb.

Still researching why. Just thinking now if there would be a way to place a piece of paper by the eye piece and take a picture of the spectrum.???

Might be to dim and the light to bright.

Hope this helps.

[Gigaah]

You think white LED spectrum is tight? the color ones are very tight. both pro and a con depending on what your after.

[inkidu]

If anyone is interested. This seems like a good deal, free shipping now >$100 till end of April, and the right parts.

From cutter a bin labeled 2700K very warm white (i have never seen a high power bin 2700K led before)

http://www.cutter.com.au/

XPEWHT-L1-8B-P4-0-01 Very Warm White

Also think that this blue works well because it replaces the nm where the warm white falls off i.e. 470-480

- XPEBLU-L1-B40-K3-0-01 Blue 470-480nm 35.2 Lumen @350mA

The cree XP-E does have a smaller die than the XP-G but if you wanted to disperse the light you might want to have

something less powerful anyways.

The # I get when you figure in free shipping is $114.80

That would be for (10) XP-E warm whites and (10) XP-E blues

Just a suggestion if anyone wants to try. Shipping is usually $12

I don't have any affiliation with the company. What makes them different is that you know the exact bin/color that you are getting

and those bins are the latest ones. Most company's don't give the exact bin #'s, unless you buy a 1000 pieces. The bin # can be important.

I personally am adding ledengin (660) deep reds and (680)

http://www.tech-led.com/data/SMB680-1100-I.pdf

to a similar XP-G and XP-E mix.

Just a suggestion if anyone wants to try led.

Hope this helps.

[rygh]

That does seem like a good deal. And knowing the bin makes a big difference.
Any problems getting things from Australia?
I have been reluctant to use them.

On other news:

1) REV-3 LED system is up and running.
I added 3 warm whites, and spread LEDs out to now cover the whole ATS. Old half-CFL is gone.
So far, looking good.

2) I seeded the screen.
Rode my bike out to the bay, grabbed some algae, and scraped it on my screen.
Not much stuck to the screen, but the few little strands that did have EXPLODED in growth.
The rest is still growing weakly as before.
So that might be the trick! I will do some serious re-seeding this weekend.

Need more time for detailed posts and results.

[inkidu]

Thanks for the reply

"Any problems getting things from Australia?" Rygh said

It did take 2 weeks but considering the distance(I'm in the USA) and they did just get off the manufacture floor (new bins)

no big deal. I have yet to test the led that I got but I have not heard one bad complaint about this company yet. (no small feat)

Gone all led. Great. You could be proof that led work for SW.

Side notes:

Will be a least 4-6 weeks before my 680 nm Marubeni led come in. My Zillionth revision

http://www.tech-led.com/data/SMB680-1100-I.pdf

I have been somewhat drawn by the high tech t-5's (been finding many advantages) but thats the past led are the future.

If anything I will know what to get, how to setup them up, where to get them, and what to avoid.

Thanks for any help.

[rygh]

You could be proof that led work for SW.
Hope so. Interestingly, the seeding seems to have done far more than all the tinkering with wavelengths.
I am wondering if that might be some LED specific thing. Perhaps what grows first changes with LED.
Sure seems like most other builds don't need seeding.


I have been somewhat drawn by the high tech t-5's..
Sacrilege. :o
I still have some T5's on my main tank, and I must admit the huge variety of colors available is rather nice.
But I always forget to replace them on time, and the sudden change in brightness is a big shock to things.
And the expense for bulbs and power is quite annoying.
I am still hoping for a bigger tank. (Working on approval from the wife).
Once that is decided, I will probably go 100% LED there as well.

[rygh]

Quick update:

I had a sudden surge of growth after I seeded, but it has subsided a bit.
It is still doing better than before though. And more importantly, the green hair seems to be winning over
the dense yellow algae. And the yellow stuff seems to grow a bit, then stop.
So seeding was definitely a help.

The other reason in increased growth I found out this week : A fish went missing.
I have a small school of blue-green chromis, and noticed it was a bit smaller than usual.
Carefully counted, and one seems to have died, and probably rotted in some rocks.
Small 2" fish, but still no nitrate spike or anything, which I am happy about.
But does explain the short term increase in growth.
I also had 2 hermit crabs die. So I hope nothing is going wrong. But did water tests, and seems fine.

My annoying bubble algae is looking a bit unhealthy now. Very excited about that. But it is not what I would call dying yet.

[inkidu]

I can't see how seeding could possible hurt. Why not try. You aren't introducing a lot of nutrients into the water with a small

amount of algae. Could you introduce something else? Sure. I regularly hide in my crawl space with a foil hat.

Quick update on my setup. Hope you don't mind Rygh

I turned off my 10 watt deep reds just because I was getting some yellow burnt areas.

Could probably increase the flow but this adjustment doesn't cost me money i.e. it is easy to unplug the driver.

So my ledengin warm whites and assortment of rebel blues are lighting the screen nicely.

I am getting good growth, should of probably went with 5 watt versions of the deep red.

Thinking, because I do have a digital aquatics controller, I should maybe try some short off and on times

with the deep reds. Still experimenting. With these results though, with the warm white and blue mix, the

cree 2700 K and 470 -480 blues that I mentioned in a previous post would be a great idea. Fairly cheap too.

Hope this helps and thanks for any help.

[rygh]

How short of a cycle are you thinking?
Have you read anything on trade offs there?
I am running 6/6/6/6 on/off/on/off right now.
Note that I only have about 30W total, so a lot less than you.

Pity about removing the deep reds.
I am convinced that the current Blue + warm_white + deep_red is much
better than the old warm_white + regular_red.

I am a bit adventurous when adding things from the wild.
In fact, when I collected the algae from the bay, I also grabbed some mussels that were
growing in the muck. Threw them in the sump, along with some muck, just for fun.
They seem to be happy, although I doubt they like the temperature.
I really believe that the more complex the biosphere the better.
That said, I am very careful quarantining from the fish store.
Like the doctors office, a good place to catch some nasty diseases.

BTW: I wonder what those govt agents in the black helicopters think of the freaky purple glow from these LED systems.

[inkidu]

Thanks for reply

The on off times I am talking about only would apply for the deep reds. (I think they are just are to powerful)

The Digital Aquatics controller comes with a power strip of 4 individual plugins

any one of which can be controlled separately. I might easily start off with a few (10-20) minutes a couple

times a day and see what happens. Aren't led great? Experimenting is half the fun

Side note

I have been very happy with the controller

it has a lot of different upgradeable paths without having a large buy in price $99

http://www.digitalaquatics.com/

bang bang

got a run they are at the door

[inkidu]

I have started to run the deep reds for an hour after the scrubber lights are on for a hour.

Great news my discus are showing signs of breeding behavior. wak a wa wa

It's working! It's working!

In two months I have brought back fish from the brink of death to breeding. You can't deny the power of the algae.

[rygh]

Very glad to hear you are having good luck.
Mine is still so-so.

My screens keep getting taken over by this fairly solid yellow algae. The pods don't touch it, and
it seems to out-compete the hair algae. And it only grows to a fairly thin thickness, then stops.
Sounds a lot like the brown/black slime, only yellow and more solid.
Maybe the LED version??
Hard to scrape off as well. Sticks between the screen and the underlying plate, and it is tough stuff.
Cleaning/seeding helps, but I can't do that forever.
As a result, filtering is not so great. Not bad. I get a lot of algae mass every week. But not a full success.

I am trying to clean screen every few days now. Will see.
I might try propping one plate up, and adding more screen, for a mini-vertical section.
Maybe it is a function of the horizontal aspect, and the mesh/plate build.

Also, I have increased feeding quite a bit, including phyto, which my few corals really like.
Unfortunately, the Aptasia like it even more. Argh. Peppermint shrimp time I think.

[SantaMonica]

Yellow is from too high a light/flow ratio. Reduce the light, or increase the flow.

[rygh]

Now I feel like an idiot.
You gave me the same advice a while back, and it helped.
But with the excitement of the new warm-white LEDs I added, I increased the amount of light, but forgot to crank back the overall time.
Then of course, I forgot your advice.
:oops:
Thanks!

[inkidu]

Don't feel bad Rygh. At least you don't have my problem I am to cheap to get a bigger pump. I use

less light instead. One day, I do want to get as much filtration as possible.

Update

Still waiting on my 680 nm led.

http://www.tech-led.com/data/SMB680-1100-I.pdf

Anyone know how to mount these?

I can't seem to find the right MCPCB.

While I am on the subject of mounting.

I have a different pair showing prespawning behavior. Although there cleaning a spot on my heater.( spot to lay eggs)

I turned it off and added a spare heater that I had. I didn't want to disrupt the mojo to much by changing things around.

I might potential have two breeding

pairs out of only five fish. Wow. Hopefully they will do the deed. I will need to separate them if they

turn out to be swingers. lol

Thanks for any help.

[rygh]

Anyone know how to mount these?
Well, other than a real PCB, my suggestion is this.

First, bend the pins upwards very slightly and very very very gently.
Just enough so that when you glue it down, they do not contact heat sink.

Second, solder wires to pins.
One wire on each side, to all 3 pins at once.

Third, glue it down to a heat sink. (Thermal epoxy )
First, put electrical tape on heat sink where the pins would hit if they are too low.
Make sure to only get glue on the heat sink part of the chip. Center square thermal pad.
It will only take a tiny amount. Hopefully that pad is slightly raised.
It says the center pad is electrically isolated, so should not be an issue there.
But if the glue hits the pins, it will short like crazy.
Alternately: Buy some thermal adhesive tape, like they use to fasten on heat sinks. Cut a small 3mm x 3mm piece,
and stick it to center thermal pad. Easier, less risk of shorting, but may not be strong enough.

Remember, no ESD protection, so take precautions. Maybe even solder in a ESD diode.

An alternative to bending pins is to build a small copper standoff, between chip and heat sink.
Probably 3mm x 3mm x 1.5mm thick.

Good luck. You will need it.
BTW: With a critical center thermal pad like that, it is basically impossible to solder by hand to a PCB anyway.

[inkidu]

Anyone looking for a led driver?

A new Meanwell led driver line just became available.

http://www.meanwell.com/product/hlg/hlg.html

Granted they are a bit large but they are very efficient (up to 94%) and have an optional

dimming function. (allows some flexibility with setup)

I have several Meanwell drivers with no complaints.

Hope this helps.

Side note

The discus keep laying eggs maybe twice a month but I think my other tank mates eat them at night.

The eggs don't last past the second day. Might need to separate breeding pair.

I have been using a water softening pillow and have only been adding RO water to top off with no water changes

for quite some time now. Maybe I should test my water(PH, Hardness, etc...) for the first time. lol

Waiting on some PH test strips right now from a ebay seller.

[inkidu]

I have been experimenting with misting systems to cool and water my tomato, pepper, and nursery plants.

And then I had a thought that the repeat cycle timer that I use to turn my water solenoid off and on (1sec. on and 59 sec. off)

might have some use with modern LED scrubber lights. (turn light on/off) It seems to me that one of the problems that I have had as well

as others is the lack of flexibility once a system is set up i.e. too much or too little light without the right amount of flow etc...

Let me first describe what a repeat cycle timer is. The timer allows the user to choose how long

the light is on and that choice plays apart in how long the light is off depending on which model is picked. (repeat cycle timer)

http://www.artisancontrols.com/pdf/4600.pdf

The benefit.

You can adjust, once the system is setup, these different times and even with very powerful lights the algae will

have time to rest between times of bright light.

Why this relates to LED lights and algae scrubbers.

1) One of the few lights that instant on/off periods have no affect on the life of the bulb.

2) There are some (this spec needs to be specified) LED drivers that can handle on/off cycles with no problem i.e. flashing lights.

3) From what I have read so far the algae doesn't care how many times the light turns on and off but the balance between photosynthesis and

respiration is of importance.

4) Allows the user to adjust timing with little effort. I use a potentiometer to adjust resistance and therefore the timing of the on/off cycles.

5) The flashing lights lets you get funky on the dance floor.

Things to consider (among the many I have yet to realize)

1) Power in and the power out. (1 amp max) My particular repeat cycle timer has AC going in and AC going out. (many different config.)

Might need a relay. guess???

2) Could there be a benefit (efficiency) in having this type of setup or is it just about flexibility.

This is all fairly intensive DIY work but not out of the realm of what is possible for a novice.

Trust me I am no expert and would appreciate any opinions and/or corrections.

Any thoughts on the use of a repeat cycle timer to turn off and on LED lights?

Hope this helps.

[Gigaah]

If your using on/off to control the light. I think generally you turn them off for more hours per day. If that isn't going to work. I think most mechanical outlet timers will on/off every 15/20 minutes. This should be fine I'd think instead if doing it more often?

[inkidu]

Thanks for the reply.

Like the idea of solid state electronics i.e. the artisan repeat cycle timer $40

The mechanical timer that I own uses those little plastic tabs to turn on and off timer,

which the manufacture only supplies a couple of.

I am sure there are other timers (solid state) but they are probably (???) less flexible and reliable than what I am suggesting.

I am suggesting to some extent smaller intervals (say 30-40 sec. every 1 min.) and strong lighting.

You could even do 3 sec. every 5 sec. the possibilities are endless and once the timer is setup it does the work with no problem.

Whether there would be some benefit in efficiency ( using strong intermittent lighting) is still a ? but there is a flexibility benefit.

Solid state electronics would IMO prove to be more reliable in these circumstances.

Might also be possible that the (slight) PH swings that can occur in FW tanks from running a scrubber would be eliminated.

I always keep in the back of my head that I am trying to run a system with very little water.

(In comparison to the large streams, rivers, and oceans found in nature)

I have read and agree with that PH swings are more detrimental than other water parameter swings (temp., hardness etc.) This being the

case for a FW discus tank.

Reasoning (IMO)

1) Adding soft water (rain) to already fairly soft water probably has little effect on hardness of the water and/or the PH of the waters with a lot of tannins.

2) Changes in water temperature after say the rainy season only stimulates them to spawn, because there is a lot of water present. (a very natural thing)
My water temperature only fluctuates 0.2 of a degree (thanks Digital Aquatics)

3) There is never much of any nutrients in Amazonian waters (the great amount of vegetation consumes any nutrients before reaching the water)
Reason I like algae scrubbers for discus.

I would assume that this is very different for SW. I have been a FW discus guy my whole life.

Hope this helps.

[rygh]

My suggestion would be to get and Arduino microcontroller, and wire up a bunch of relays.
(Or power transistors, if you really care about solid state)
Look at sparkfun.
I got really tired of all the individual timers and everything. Plus, it is cheaper if you have a lot.
Anyone that can wire up an LED system can probably handle the micro.
Plus, you can do a lot more, for example, I have a temperature sensor that can turn on my chiller.
(Soon to be changed, to turning on a fan across the ATS screen)
Bonus: It gives you an endless opportunity of fiddling... :-)

---

Interesting about the new Mean Well drivers.
But I still really prefer the simpler single-chain, specific current devices.
No risks with current sharing. Easier setup.

---

I am not confident the algae will be happy with a 15 minute cycle.
The algae does more that "rest". The organic process changes a lot. Supposedly, most of the growth occurs when it is dark, while energy conversion occurs
when it is light. It takes a while for cells to switch between those processes. Likely far more than 15 minutes.

I did try a 12 hour cycle, and went back to a 24 hour cycle. My impression was that the 24 hour cycle was slightly better, but I changed other things as well,
so I do not consider that very useful data.

[inkidu]

Thanks for the reply.

Will look into Arduino microcontroller I have heard about them and I like the concept (open source, many app.)

Good news I just got this email from Marubeni about my 680 nm led. (total of 10)

http://www.tech-led.com/data/SMB680-1100-I.pdf


Hi,

We are shipping your order this week, as early as tomorrow. We will send you a notification when we finish shipping arrangement.


Kumiko Quanrud


Was thinking about the Meanwell HLG-100-48

http://www.meanwell.com/search/hlg-100/default.htm

I am basing my decision on the ( Forward Current - Forward Voltage graph)

At the low setting of 1.2 amps the forward current will be approx. 3 volts (10*3 = 30 within the constant current region)

At the high setting of 2.0 amps the forward current will be approx. 4 volts (10*4 = 40 within the constant current region)

Both settings seem to lie in the sweet spot of the LED ( judging by the Forward Current - Forward Voltage graph)

Anyone have opinion?

Am I missing something?

Thanks for any help.

Side note.

Seems that I remember about reading that the change between respiration and photosynthesis happens very quickly.(with algae)

The resource

http://www.life.illinois.edu/govindjee/photosynBook/

Whether lab results always carry over to the real world is another thing.

[rygh]

I like to run at 750mA.
So I would have to make 2 chains, and run at 1500mA.
But with 2 chains, if I do something wrong (possible), and one chain stops, then the other
chain gets the full 1500, and fries.
Plus, chains really need to be identical or current will be unbalanced.

Plus, I like to check it with different configurations of LEDs.
With the simpler ones, I just have one chain, and a very wide range of voltage, which means I can
hook up pretty much whatever I want, without thinking about it or tweaking any settings.

On the other hand, it means more drivers, so more expensive.

[inkidu]

Just an update on what is going on in the led world for anyone interested.

The Cree XP-G warm white (2800K) led are becoming more available (sometimes a manufacture comes out with cool whites first then as time passes they release other types i.e. warm white, neutral, outdoor etc....)
XPGWHT-L1-8B1-Q4-0-01 Warm White 8B (IMO would be a good choice)
http://www.cutter.com.au/products.php?cat=Cree+XPG (scroll to the bottom)

The Cree XM-L will be out in couple months but will probably need to wait on the warm whites.

I mention these things because the tech changes so quickly and the new led are significantly better.

The new Cree XM-L (single die)can handle 2 mA and produces about as much as a 60 watt bulb i.e. 750 lumen.

That is 7 watt led can replace a 60 watt incandescent bulb.

WOW

http://ledinside.com/News_cree_LED_20100513

Those and a MEANWELL HLG-100-48 could make quite a big setup.

http://www.meanwell.com/search/hlg-100/default.htm

[SantaMonica]

I like that lens at the bottom of the Cutter page.

[inkidu]

More LED news.

A cheap and efficient deep red LED from Osram

LH W5AM-1T3T-1-L-Z

http://www.mouser.com/ProductDetail/Osr ... NWtQ%3d%3d (http://www.mouser.com/ProductDetail/Osram-Opto-Semiconductor/LH-W5AM-1T3T-1-L-Z/?qs=dzkBnbp7AzQjkL7H45NWtQ%3d%3d)

Still need to get my feet wet on mounting these LED emitters to boards but it is nice to see some more options become available.

Hope this helps.

[SantaMonica]

I think the led scrubber will be feasible soon... just got to get the initial price down.

[rygh]

Well, if you are willing to go with cheap ebay warm white + red LEDs, the cost is down.

You can buy those 3W LEDs for less than $2 with shipping off ebay.
So a 35W system ends up only about $60 or so.
Get some volume discounts, and a cheaper power supply, and that is $50 or less.
Which is about the same as a good T5 setup.

Sure, not as efficient as our fancy deep-red thoughts (maybe), but good enough.

[rygh]

Experimental results:
I did this mostly as curiosity, and to test my new main-tank LED sets.
But very relevent on how it shows very clearly how the wrong spectrum can render an ATS useless.

The left side is driven by my new main tank LED set. 2 x XP-G Cool white + XP-E Royal Blue.
A fairly common 10,000K-12,000K equivalent bluish light. Looks really nice in my tank.
The right side is driven by my algae-specialized LED set. 2 x Deep Red + 1 x Blue.

This picture is the lighting. Obvious color difference.
[attachment=1:27gtj9nq]experiment_1.jpg[/attachment:27gtj9nq]

This other picture is the result. You can clearly see the lack of algae growth on the left.
Unfortunately I did not take a pic earlier, but it turned a rather ugly dark brown. I scraped that off,
and basically very little grew there afterwards.
Note: Resolutions/angles are different, but that line of good/bad lines up with LEDs.
[attachment=0:27gtj9nq]experiment_2.jpg[/attachment:27gtj9nq]

[SantaMonica]

Yes a dark brown growth is a symptom of the lighting not being strong enough relative to the nutrient levels. Thus, dark growth to start, until nutrients come down, then lighter growth. But your red led's deliver more lighting for growth, and thus the green growth is able to "power through" and grow right away.

[inkidu]

More LED news.

Notice the second paragraph. (warm white LED)

http://www.led-professional.com/technol ... in-the-lab (http://www.led-professional.com/technology/light-generation/epistar-red-orange-led-chips-set-efficacy-record-in-the-lab)

In case the link breaks search this.

Epistar Red-Orange LED Chips Set Efficacy Record in the Lab

Still in development but it is encouraging to see the development of the red led tech ( AlGaInP, AlGaAs etc...) instead of mainly "blue/cool white" led.

Hope this helps

[rygh]

Yes, interesting news.
Fortunately for our main tanks, we like the current extra blue used in the conversion to warm.
But for lights around the house, sounds great.

[inkidu]

More led news.

The new Cree XM-L cool white led are shipping Mid Nov..

http://cutter.com.au/products.php?cat=Cree+XM-L

These are the latest and greatest cool white led available.

Notice the maximum drive current: 3000 mA WOW! (this # for a single die led is very large)

Times are changing.

Hope this helps.

[rygh]

An amusing note:
Because of the ongoing build for my new tank, I left the poor white/blue section as-is.
The funny thing is, I am getting a bit of Coraline algae growth there.
Guess Coraline likes the blue.

Note that with 1/3 of the filter not doing well, I definitely get more nuisance algae in the main tank.

[regie]

I will be using that one if I got my tank bigger than what I use now. But for now, I am using an ordinary light but I plan to replace it by a led light bulb. My friend explained to me that these lights consume less electricity than the other light bulbs.

[Parasurama]

Well this is my first time to read an LED light bulb can grow algae. I am using in growing the algae is the old steps but now there are modern LED scrubber light to algae. Now I decided to change my old light bulb in my tank in much good result I will replace a LED light bulb so my algae can grow regularly.

[SantaMonica]

Remember that LED's are for experimenting only... do not use them if you must have filtering that works.

[inkidu]

More LED news.

Cree Launches Highest-Performance Warm & Neutral White Lighting-Class LED

http://www.cree.com/press/press_detail.asp?i=1299160266032

I order some neutral and some warm to see how well they work.

Hope this helps.

[SantaMonica]

Is anyone making a power led-wall, with high power and consistent intensity across a large surface? I've seen low power walls for signage backlighting, but nothing meant to project high power.

[Mania]

Just finalising scrubber design now, it's currently planned to have panels of LEDs attached to the wall the tank sits on top of (it's a 220g in-wall tank) to light a single sided vertical scrubber (75cm * 40cm).

Will certainly let you know how it goes.

[Floyd R Turbo]

I talked to my regional Cree rep about XM-Ls vs XP-Gs. XM-Ls just might be too intense. You would have to diffuse them so much that it might just make more sense to get a couple XP-Gs and diffuse them only a little.

[rygh]

The micro-diamond diffusion grating from Lowes/Home depot works great for diffusing LEDs.
Basically, it is an acrylic sheet, with little pointy diamond shapes on one side.
If memory serves, about $10 for a 2' x 4' sheet.
Put it about half way in between LEDs and screen.

It scatters light well, with fairly low loss.
I do not see any point source issues on my scrubber after using it.

Since it is flat on one side, you can easily glue two together if you want.

---

Interesting about the new XP-G warm whites.
Although not sure it really makes much difference. From my experimenting, key is to have lots of red.
My best results (although limited not-very-scientific testing) was 2 x red + 1 x blue + 1 x warm white.

---

Someone PM's me with some interesting reading:
http://en.phyco.org/wiki/Algal_Light_Requirements

[inkidu]

More LED news.

Just something someone might want to try out.
light coverage is rather square which would be good for our purposes
and mounting system is easy to use/update (watch video)
Notice also the custom spectra choices and the high IP rating

http://www.theledstore.com.au/product_p/ls0017.htm

As far as that make I have no experience.

Hope this helps.

[salty joe]

Just curious-how are the LED scrubbers doing these days?

[rygh]

Just curious-how are the LED scrubbers doing these days?

For me, I am happy.
I have some pictures here, about half way down:
viewtopic.php?f=3&t=645&start=30 (http://www.algaescrubber.net/forums/viewtopic.php?f=3&t=645&start=30)

I have become really lazy on water changes. :)

I do have a bit of hair algae in the main tank.
But recently, it is looking a bit less healthy, so I am encouraged by that.

[inkidu]

I have become really lazy on water changes.

Happy to see things going well.
LED and a algae scrubber IMO are all about less maintenance/work. (I like to say less expense???) Time will tell.
There is the engineering fun behind it of course also.

More led news

Affordable warm white led on the horizon? Hopefully

http://bridgelux.com/media-center/categ ... -releases/ (http://bridgelux.com/media-center/category/press-releases/)

Hope this helps.

[rygh]

I have become really lazy on water changes.

Happy to see things going well.
LED and a algae scrubber IMO are all about less maintenance/work. (I like to say less expense???) Time will tell.
There is the engineering fun behind it of course also.

More led news

Affordable warm white led on the horizon? Hopefully

http://bridgelux.com/media-center/categ ... -releases/ (http://bridgelux.com/media-center/category/press-releases/)

Hope this helps.

Interesting! It would be great if the price of WW LEDs drops a lot.
But for turf scrubbers - ehh - Not a fan of warm white anyway. Red+Blue!

[SantaMonica]

So what is the current cost of 100 watts of red+blue, if it's 1000 pieces of 100mw each

[rygh]

So what is the current cost of 100 watts of red+blue, if it's 1000 pieces of 100mw each

Not sure about 100mw LEDS. Assume you meant the cheap 5mm little ones.
I also assume you mean buying mass-bulk, since a DIY hobbyist would go insane doing that many pieces.
A quick check, 100W looks like about $60.
Looks like you can get 20mA (50mw), red 5mm LEDs for about $0.02 to $0.04, so lets call it 3 cents.

Note that those little ones are not as efficient as the bigger ones.

[SantaMonica]

I think that distribution is more important than efficiency. If you think about how much of a difference moving a cfl a few inches makes, think about how the LED intensity changes by moving outside of it's beam cone will alter growth. And of course it would be done with wave soldering.

With 1000 per side, you can lose 20 or 30 and not really even notice it.

[rygh]

I think that distribution is more important than efficiency. If you think about how much of a difference moving a cfl a few inches makes, think about how the LED intensity changes by moving outside of it's beam cone will alter growth. And of course it would be done with wave soldering.

With 1000 per side, you can lose 20 or 30 and not really even notice it.

Too bad there is no real volume for these things, it might be fun to really build a production board.

Light distribution issues can be fixed with a very cheap diffusion grating. Available at lowes, OSH, etc, about $10 for a 2x4 sheet.
I definitely use one, and as such, have no issues.

Interestingly, I think some of the standard CFL systems I see here would also benefit from some cheap diffusion grating in
front, and better reflectors in back. Optics seems to be a neglected part of many builds.

[inkidu]

More led news

High efficiency and high output royal blue led from Cree.
http://www.cree.com/products/pdf/XLampXT-E_ROY.pdf

Hope this helps.

[rygh]

TWO YEARS !!!!
8-)

Amazingly, it has been two years now since I started my first LED based scrubber.
I started this thread the next day.

I did have to upgrade my scrubber - when I changed to a larger tank.
And I switched to a true vertical double sided.
BUT: Other than the additional ones, the LEDs are the same ones I started with.
Nice not to have to change bulbs, and I think I saved a lot of energy in those 2 years.

Overall, still very happy with it.
Nitrates = 0
Phophates = 0
No nuisance hair alage.

I do still run a skimmer, and plan to long term.
I also run a bit of carbon, and I do the occasional water change.
But a lot less chems/water changes than other reef friends I know.

Tank is still in slow progress though.
Only a couple of SPS frags.

[SantaMonica]

Lots of room for the frags to take over.