time to upgrade my LED lighting - The Planted Tank Forum
 
LinkBack Thread Tools Display Modes
post #1 of 9 (permalink) Old 04-17-2016, 12:03 PM Thread Starter
Algae Grower
 
PTrader: (0/0%)
Join Date: Apr 2016
Posts: 8
time to upgrade my LED lighting

several years ago I built a LED light fitting for my tank which is 5'x2'x2'.

control is with the jarduino aquarium controller which looking at the code puts my build back in 2012

it looks like some things never change as the cost of the LED's was then as it is now a significant cost of the build so I went down an experimental route with 50w cob units that were just appearing at the time.

the plan was to test things out as a proof of concept and once all was hopefully working to build a better system....... 4 years later it worked so well that I figured I should get around to the final version!

currently I have 5 clusters containing 1x 50w 6000k, 1x10w rgb ad 1x 1w blue

so ~ 300W in theory, the rgb was added as there was little data on the cob units so I wanted to be able to tune the colour if needed though in the end it was fine and these just run as white. the blue is moonlighting.

one of the main changes is going to be heatsinks. at the moment the cob's are on small active cooled heatsinks but I have managed to acquire a load of 12"x8" heatsinks which I will join together to make one 6'x12" heatsink forming the basis of the lighting unit.

this should be more than enough to be passively cooled.

when it comes down to emitters I am thinking of Cree xm-l2 just because I have found a deal with them at 1.52 each (under $3)

I have looked at vera and luxeon but price wise the xm-l2 seems hard to beat. any thoughts??

my current plan will be 7 clusters containing:
1x 50w COB
5x 10w xm-l2
1x 10w RGB (moonlights)

now I know this sounds like a crazy amount of light but hear me out

I currently have 300W being driven hard and its about right - could use a little bit more but its almost there.

although I would have 750w with the new arrangement I would be under driving it so less stress on the emitters, less heat and ultimately about half the overall total.

for drivers I am still not sure. I have enough to cover all this but may change to the LDD1500 as the packaging would simplify things a bit.

so any thoughts?
opentoideas is offline  
Sponsored Links
Advertisement
 
post #2 of 9 (permalink) Old 04-17-2016, 09:13 PM Thread Starter
Algae Grower
 
PTrader: (0/0%)
Join Date: Apr 2016
Posts: 8
now I remember why I have avoided this for 4 years....

just spent the afternoon looking at what advancements have been made in arduino based controllers including reading through the whole iAqua thread

now trying to decide which way to go. really like the idea of 12 bit pwm and android integration and hate to say that the bluefish mini may be the way to go.

I need to keep looking as hopefully someone has done an open source project. while a can work code on the arduino I am nowhere near to writing it out from scratch.

if anyone knows of such a project please let me know as I am still looking
opentoideas is offline  
post #3 of 9 (permalink) Old 04-19-2016, 07:30 PM Thread Starter
Algae Grower
 
PTrader: (0/0%)
Join Date: Apr 2016
Posts: 8
Crickey - its lonely in here

Oh well will keep talking to myself for a while

Not got everything figured out yet but heading in a direction I think....

For controller much as I prefer to DIY the Bluefish mini seems too well suited to ignore. for the short term I will have a play loading iAqua instead of jarduino that I am using. Once I start playing with that I may change my mind but I don't see myself using all the extra features so will most likely swap this out for Bluefish down the line.

Driver wise I think I vaguely remember looking at the LDD series when I first built the lighting but they were new and not really available at the time. I went through a load of rubbish drivers that had a habit of blowing themselves up (and the arduino!) for no apparent reason. Although I have more than enough drivers to do the new layout they are dubious quality so probably best replaced.

Given the LDD is now both easily available and dirt cheap I think they will be the way to go - not to mention if I do end up with the Bluefish then changing them now makes that a simple change to make later.

Emitters are my last question mark (and the first for that matter)

Where I am struggling is looking for any up to date information. most of the information I can find is from 2011-2013. While there is loads of controller information DIY builds seem to have disappeared. does everyone buy commercial fixtures and mod them?

4 years ago emitters were changing and improving on an almost daily basis. has nothing changed since then?

I can see that a number of newer emitters are higher current than drivers like the LDD can cope with. Is this why older emitters are still used? have I managed to miss all of the great lighting builds done in the last few years??

I noticed that the LDD-H could be run in parallel but was limited to ~1400mA I don't suppose anyone knows if there is any similar restriction on the LDD-L?? I feel an experiment coming on
opentoideas is offline  
 
post #4 of 9 (permalink) Old 04-19-2016, 10:10 PM
Planted Member
 
PTrader: (0/0%)
Join Date: Jan 2014
Location: Federal Way, WA
Posts: 215
FWIW I have been happy with the Bluefish Mini, but you are somewhat limited by it compared to writing your own code - or even compared to using a more flexible controller like an Apex. You've only got a set number of ramps available: Night -> Morning, Morning -> Mid-Day, Mid-Day -> Evening, and Evening - Night. But for what it is, it was pretty foolproof to setup and has worked flawlessly every day since I fired it up.

-Justin
jrygel is offline  
post #5 of 9 (permalink) Old 04-20-2016, 06:44 PM Thread Starter
Algae Grower
 
PTrader: (0/0%)
Join Date: Apr 2016
Posts: 8
Hi Justin,

Thanks. That sounds like exactly what I hoped.

I already have a Jarduino controller and for all the things it can do......in the last 4 years I have used it to dim the lights and even once or twice changed the clock when it changed to summertime.

For all the very clever things can be done with microcomputers I have rarely seen diy efforts that had more than a passing thaught given to fault tolerance and to be fair to those that come up with these great controllers they are limited in so many ways it's not surprising.

I went with the Jarduino because it was one of the first to do a sunrise/sunset fade on the lights and that is a great idea but I don't feel comfortable leaving anything much else automated on a tank.

That's one of the reasons the Bluefish is appealing it's so simple and user friendly without all the bells and whistles hopefully it just does what it does and does it well.

Glad to hear it's working well for you. It is good to hear it's all its advertised as.
opentoideas is offline  
post #6 of 9 (permalink) Old 04-21-2016, 05:35 AM
Planted Tank Guru
 
PTrader: (0/0%)
Join Date: Jun 2013
Location: WI
Posts: 11,128
The de-facto champs for all out CRI and efficiency are Bridgelux Vero Decor series..
Sets of ww/neutral or cool white is all that is "really" necessary.
high efficiency, low price but w/ some, as you know drive current liabilities..
They can far exceed the output LDD's (up to 4200mA )can deliver in many cases.
Second downfall is thermal management of such large yet dense arrays of chips on board..
Passive cooling at high drive currents is "problematic"
Third downfall is getting decent lenses. Proper placement and even for a deepish tank not really necessary..

All of the above can be remedied by $'s though. But kind of defeats the cheap chip thing.

One of the reason small emitters are still popular. And of course the VERY cheap prices of cheap 3W chips, even if you need 2 or 3 to one compared to name brand emitters.

Since aquariums are more of an afterthought in the manuf. world all have compromises. The large COBS are designed around OEM canned lighting uses.

On the small end little .5W smd's (much harder to work w/ for DIY) are exceeding 200lumens/watt and rising.
These are also targeted to "architectural" or home use..

But for us the large dense arrays of small emitters on circuit boards are not easily worked with.
One chip I would die for to have on a 4' board are the Yuji smd5630's that use a violet "base" emitter and phosphor pack for >95CRI..covering ALL wavelenths of visible light for the most part. No cyan shortage, no "violet" hole..

The point is, since none of these things takes DIY and aquariums seriously the "best for our use" will be some accident in creation..

I don't even want to get into the individually "addressed" diode strips which could actually allow "zoning" in an aquarium.
Currently DMX based for stage and advertising ect.

There are probably a dozen "innovations" that have crashed in prices and are, in some form or another usable but just hasn't hit any type of density...or has been well thought out.

I've discussed and seen working large , greater than 3A custom drivers ala LDD style ..and Meanwell and inventronics and make large ones as well but mostly as ps/drive "units" w/ a price to match.

On a more specific note "if" you can get them for a cheap enough price (or even get them..) pairing this:
BXRC-56G4000-F-24 5600K 90 CRI 113lumens/watt at @1050mA
28.8v and 30w/chip w/ a low k chip..
BXRC-25E4000-F-24 80cri 2500k 118 L/watt at 1050mA and 31W/chip
Both run full on effective temp of approx 4050k w/ a good deal of red and even a bit of a cyan boost is pretty much all you need.
Color to taste..
Each cluster will run approx. $30. Add 2 LDD's for $16 and you get $46 for 7000 Lumens per cluster..
Equiv. to 70W of T5's in a tiny package..(on a lumen basis, PAR is different, as is "delivery" geometry)
for your tank you'd probably want 6 "sets"..$276 plus asst. parts and power supply..

Oh and easily pushed up w/ 1500mA drivers..losing a bit in lumens/watt but gaining watts..pushing the 2 chips up to 3 t5 equiv. approx..w/ no extra cost w/ the proper ps. Adds 10W each to the chips..500w ps guesstimate..at 36V but only a ballpark figure..
Add a RGB to each set for fun...

Oh and just because a chip can "take it" doesn't necessarily mean you should give it..

http://www.bridgelux.com/sites/defau...%20Rev%20C.pdf
http://www.cree.com/~/media/Files/Cr.../XLampXML2.pdf
you asked..


Oh and you can modify a TC-420 controller to run the whole thing.. $25-$30 usually... 5 channels..
It would take 19 CREE l2's to equal 1 cluster (375-ish lumens/1000mA) .. though at your price.. Under $56.. What color temp btw..?

you would need to push them to 3000mA to get close to 10W out..and may hit 700 lumens/chip but good luck w/ that..
The cost would be around $30 (10 chips) for a "cluster equiv" at 1000mA vs 3000mA

down and dirty calcs..and throwing out the 50W-er all together..
The Bridgelux at 1500mA is almost 5000 lumens/chip x12.. Grand total of 60,000 lumens over your tank..
20 4ft led shop lights..

"A man with a watch knows what time it is. A man with two watches is never sure."

Last edited by jeffkrol; 04-21-2016 at 06:21 AM. Reason: edit
jeffkrol is online now  
post #7 of 9 (permalink) Old 04-23-2016, 06:26 PM Thread Starter
Algae Grower
 
PTrader: (0/0%)
Join Date: Apr 2016
Posts: 8
Quote:
Originally Posted by jeffkrol View Post
The de-facto champs for all out CRI and efficiency are Bridgelux Vero Decor series..

//

The point is, since none of these things takes DIY and aquariums seriously the "best for our use" will be some accident in creation..

//

On a more specific note "if" you can get them for a cheap enough price (or even get them..) pairing this:
BXRC-56G4000-F-24 5600K 90 CRI 113lumens/watt at @1050mA
28.8v and 30w/chip w/ a low k chip..
BXRC-25E4000-F-24 80cri 2500k 118 L/watt at 1050mA and 31W/chip

//

you asked..

//

It would take 19 CREE l2's to equal 1 cluster (375-ish lumens/1000mA) .. though at your price.. Under $56.. What color temp btw..?

you would need to push them to 3000mA to get close to 10W out..and may hit 700 lumens/chip but good luck w/ that..
The cost would be around $30 (10 chips) for a "cluster equiv" at 1000mA vs 3000mA

I did indeed ask

great information and references

If you dont mind I have a few queries though...

while CRI is a good indication its only a measure of I think 8 points on the chart?

when you compare the spectrum charts for both the vero and Xm-l2 there is little difference I can see. have I missed something?

I dont quite agree with your sums either. from the data sheet you linked the cree (U2 3C bin I am looking at) are 580Lm at 1500mA

price wise for the cost of the 2 vero I could get 14 of the xm-l2 so the output works out near the same for the money.

this is one of the reasons I am considering them. I know I am being unfair to the vero as at 1500mA it would increase 30% however you also make a good point about heat.

a single emitter like the vero has to dump a lot of heat several smaller stars have a much larger total surface area.

cost obviously comes into it and I am looking to keep the cost down. I already have the controller and I have a huge area of heatsinks to play with. my only costs are the drivers and emitters. the drivers dont change with any option so its only the emitter cost I am looking at.

if I went with the vero its 130 if I reuse my existing cob's and added in 35 xm-l2 it would only cost 45

for less than the cost of the vero I could buy all the drivers and add in a pile of xm-l2

from what I can tell on the spectrum charts of these different emitters there isnt a massive difference in the spectrum between them

I am just surprised that 4 years on its pretty much the same LED's that are at the top of the heap.

I have seen some of the Yugi offerings and they have some nice products but still a premium price. fingers crossed whatever they are doing catches on as it looks like in a few more years the ideal emitter may be available at the right price point if the demand is out there in the commercial world.

the more I look the more I think I will just treat this as a chance to finalise the build with the upgraded heatsinks and framework. change to the LDD's to sort out the drivers and supplement the emitters I have.

that would then give me the foundation to upgrade the emitters at a later date if and when a real improvement is possible.
opentoideas is offline  
post #8 of 9 (permalink) Old 04-23-2016, 11:03 PM
Planted Tank Guru
 
PTrader: (0/0%)
Join Date: Jun 2013
Location: WI
Posts: 11,128
There is more than one CRI "metric" . 8 color swatches was not enough and failed miserably w/ LED's so CRI can be tricky.
also it is "weighted" depending on what standard you use.. Less than 4000 or so and it is measured against an incandescent bulb. higher color temp and you go to d65 d50 ect.. More "sunlight" like CRI.

Current CRI is using 8 plus 6


R9 is what I consider the critical one..
Always look at the spectrum charts..
Always look for less "dip" in cyan and more tail end in red..

none besides the Yuji have ANY "purple"...

As to costs.. yep understand that well.. but I do like to know how much the high end costs vs the low and set my compromises from there..
Technically there is no need for any CRI talk..Radion x15 FW charts at like 17..

Except.. the higher the CRI the more the red in a white w/ LED's.. pretty standard

Quote:
a single emitter like the vero has to dump a lot of heat several smaller stars have a much larger total surface area.
Yes but some of that depends on design as well.. If you can't get the heat out of the chip heatsink size doesn't matter.. Doesn't really apply just an FYI.
Most Vero builds use fans. There are "made for passive cooling" heat sinks for veros but again, that cost thing..
Recently found what I consider a better chip than the VERo's Unfortunately the cost is much more as well..
Luminus Devices "studio" series..
CVM-14-56-95-36-AC00-F2-2
Quote:
Flux @ Current/Temperature - Test 2800 lm (Typ)
Current - Test 1A
Temperature - Test 100C
Voltage - Forward (Vf) (Typ) 36V
Lumens/Watt @ Current - Test 78 lm/W
Current - Max 1A
CRI (Color Rendering Index) 95
Not sure why I like it better though.. weaker, less efficient, and more expensive..
But sure do want to pair up a 3030K /5600k .
Oddly the only 2 K's they make in that series..
Then they have this one..
http://www.mouser.com/Search/Product...4309536AA00F23
95CRI a bit weaker BUT 1/3rd the price..
Catch to all these chips.. 36V V(f) Using ldds and those dropping a few vots makes standard 36V ps's iffy..

Currenty facing my own analysis paralysis...

for fun:
Check out the subtile difference between the 90 and 97CRI 3000K diode.. Slight decrease in green and a slight increase in "cyan"..

cleaned forfun:

"A man with a watch knows what time it is. A man with two watches is never sure."

Last edited by jeffkrol; 04-24-2016 at 12:20 AM. Reason: edit
jeffkrol is online now  
post #9 of 9 (permalink) Old 05-07-2016, 09:20 PM Thread Starter
Algae Grower
 
PTrader: (0/0%)
Join Date: Apr 2016
Posts: 8
Quote:
Originally Posted by jeffkrol View Post
for fun:
Check out the subtile difference between the 90 and 97CRI 3000K diode.. Slight decrease in green and a slight increase in "cyan"..

cleaned forfun:
see this is the bit I cant figure out. does it REALLY make any real world difference?

ok on a graph there is a difference but thats a marketing tool, but in the real world and in our case where I suspect that any minor variations would be gone after the first couple of inches of water penetration is there any actual difference? I just dont know but from what I can tell on the graphs the differences between emitters are so slight I suspect that the difference is far less than the cost difference.

A PAR meter would be interesting for comparison but again I have not seen any real independent tests and most of the "tests" done have been by the systems builder who naturally has a vested interest in justifying the expense of what they have built. not that they wont be accurate but budget builds like mine rarely would have access to a PAR meter and I would be fascinated to see if there was any real difference between the best and cheapest chips out there when directly compared in the same setup.

as to your own paralysis remember that a good 36V PSU will have a voltage adjuster and with the 36V Meanwell PSU I have it can be raised to 40V so the 3V drop on the LDD would still let you run a Vf of 36V if needed.

I am also having a play at the moment with LDD-H and LDD-L to see how they actually perform as I suspect the spec sheets are a bit conservative as my first test with a LDD-H had it pushing an output at 35.3V from a 36V input which is nowhere near the 3V loss I was expecting.
opentoideas is offline  
Reply

Tags
None

Quick Reply
Message:
Options

Register Now



In order to be able to post messages on the The Planted Tank Forum forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.

User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.

Password:


Confirm Password:
Email Address
Please enter a valid email address for yourself.

Email Address:
OR

Log-in










Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page
Display Modes
Linear Mode Linear Mode



Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

 
For the best viewing experience please update your browser to Google Chrome