LED Custom 140

[DIYer]

To all those who want to DIY,.. and LED Girl who wants your money...

To all those who want to DIY,.. and LED Girl who wants your money...

Supra's awesome DIY LED panel is the reason for this threads existence, and thus should be much more then a "side note" during conversation here. If this were my thread i would tell you to take a hike and go hawk your inferior product vs what Supra made himself in the vendor section where people look to buy not DIY. To repeatedly hawk your product in someone elses thread is rude.

I've DIYed my own LED grow lights now too and neither me nor Supra had to pay to "figuring out the correct ratios" bla bla bla, because Knna did it for us for free based on the exact area we intended to grow in. Let me see another LED grow light seller do that. And unlike when i asked LEDgirl months ago what brand of LED's were used, there spec's, and so on, Knna actually answered my questions without getting all defensive and basically tell me to F-off like LEDGirl did. Yes that's right, that actually happened and when i reported it to the mods they told me it wasn't the first time she's had that complaint lodged against her.

Roortokeage my DIY'in man, if you want to do your own HERE'S HOW. Knna shows us the way and tells us what LED's to buy, there specs, and everything in that thread. Not to mention you'll learn a lot more about LED light reading that thread then any seller of LED grow lights will ever tell you, or could. It might not be cheaper to DIY in the end vs some products out there, but it wont cost more and you WILL have better lights in the end (not to mention better distribution of that light) and that's really what matters. Look at Knna's panels on page 2 of that How to build your DIY LED array thread. Having all the lights in one tight spot is old school, it's how narrow minded manufacturers still arrange lights for the sake of convenience, anyone who can solder can DIY a better product for the same price and if you just take your time and do the job right like Supra did you wont need a 3 year warranty since all LED's easily last for much longer then that.

Sorry to go here Supra, but as i said to you before privately, i feel the truth needed to be said. I'm just trying to point those who care in the right direction to DIY there own, and im sure manufactures of LED grow lights will still sell plenty to those who don't care to have the best of the best in the end

 

[SupraSPL]

I'm glad to hear all points of view and I am glad to see LEDs succeeding commercially and on the DIY front. I think the design differences between Hydro-Grow and KNNA style are well established on p2. The fact is, there are a lot of differences, which may or may not matter to any particular grower.

I threw one of my modules in a small vegging cab and ran it on a single driver on 18/6. Input power was at 30 watts. The growth came on fast and strong, with nice tight nodes. I actually had to turn the driver down and now reads 13 watts. Its nice to have a "growth speed knob" in the veg chamber. I am very excited about the vegetative possibilities for LED, especially with the CREE XP-G R5 bins and the new 660nm I have heard rumors about.

When it comes to flowering, that depends on the grower's situation. I have had the privilege to see elite cuts grown under 1000watt SuperHPS and they are no joke, and that is the competition ultimately. Cheap and easy to set up lamp and grow room. Currently, if a grower with those needs asked me to build him/her an LED I would point them to Hydro-Grow because I wouldn't want to spend the time to build it.

But I have to tinker and if it were my grow I would build a 500w LED with top bins, heatsink fins outside the chamber and adjustable drivers. It's a case of to each his own. In my mind every last watt counts, and we scale it up from there.

 

[tribe]

Go away LEDgirl. Too scared people will see how its possible to DIY it and not give you all their money?? If people want to buy an LED rig they'll look for you, this is the online equivalent of "cold-calling" sales. People are not reading a thread entitled 'led custom 140' because they want to buy one of your LED setups. Apart from maybe a token gesture at the end to "cover yourself" you just dis people and big yourself up. Go get laid.

Shakes head. Some people just don't understand the joy of building your own, the fun you can have with making something yourself and solving all the puzzles and challenges and not to mention the experience gained from it and that great feeling when it works! Why are we all growing our own ganja and not just buying it off the street...?? Similar principle. Sorry about that SupraSPL, but what a place to advertise, I will edit away if you ask.

Full respect to you SupraSPL, I was thinking that a homemade LED unit could be made, but know little about it and before I did a huge read up wanted to see someone who's done it before. Nice setup and nice handywork there! The heatsink idea from a cpu? Expanding on that, do you think it would be possible or worth watercooling it? Thanks for the inspiration to DIY LED. The variable wattage power knob dial majiggy you have on the veg cab is a fantastic idea! Rep+

I will certainly be attempting this myself too as soon as I my first grow under my belt... and I will be sure to spread on my knowledge as you have (if only to wind up LEDgirl)...

 

[SupraSPL]

Haha its all good Tribe all opinions are welcome here including yours. There are lots of differences between the lamp designs that can be debated. I am in the process of collecting all the parts for another build and my wallet is definitely feeling the pinch this time lol. I can respect how the HGL lamp design found a way to make usable LED available to almost anyone and still make a profit. On the other hand the DIY lamp can maximize efficiency and these projects are a ton of fun to build as you pointed out.

The heatsinks from CPU coolers can work very well but generally they do not have enough surface area to support very many watts when passively cooled (~10 watts each). The heatsinks I used are very similar to the ones from Heatsink USA LLC and you can get them from their website or eBay.com seller barrredboss. The conversion efficiency or radiometric efficiency of the LED dictates how much heatsink surface area you would need per watt. As they improve the efficiency the less heatsink we will need per watt.

As far as water cooling goes, KNNA has always recommended the passively cooled design thanks to its simplicity/efficiency and reliability versus adding fans. However active cooling can allow for more focused light or more leds closer together. That normally leads to reduced efficiency but in some cases it may help to maximize a small space. I think Pinstripe's 22w PC grow is a good example.

The attached spreadsheet can calculate the surface area a heatsink with a few easy measurements. XLS is not a valid file type for ICMAG so you have to rename it to .XLS when you download it. It comes in handy for quickly calculating heatsinks on eBay and already has the data entered for the two best the Heatsink USA LLC site. The sheet is .XLS but it will open in openoffice or google spreadsheets with no problem.

 

[DIYer]

Hey Sup whats up! As for buyin' or DIYin' i only got one thing to say. WHATS MY NAME BITCH!! ..but no i love the buyers too, if it weren't for them id be out of business. The world needs both, i just wish more of the products out there contained the parts like we both got from Knna ya know. I hate when people sell an inferior product just to make more profit. I could do the same thing with what i sell but why not just make a better product (like we did) clear a little le$$ per unit, but sell more units which would prob work out to about the same in the end and then you can go to bed at night knowing people are growing more bud cuz of your superior product.

I've made 2 complete LED setups so far, im growing in them both right now, and ill be posting grow journals soon. One is in my 26.5" tall LED PC Grow Case design (2 plants) and the other in a completely new device i designed and built that grows two 2ft tall plants as well and gets this,... are you ready for this,.. the whole grow box slides under any bed!
I'm not gonna lie, i plan to sell it too just like LEDGirl is her lights, but i wont be hawking it to those looking to DIY one of there own. But maybe im no better then LEDGirl saying all this now,... well except my lights will be better, Ha! ...and i wont turn my nose up to people who ask how to DIY there own. No im just hella excited about it and everything LED ya know. If you want PM me and ill send you a link to a private journal i got going thus far so you can see it in action.

Take it easy fellow do it yourself'er!


~peace~

 

[SupraSPL]

Hey DIYer I'll have to check them out. Did you post on IC or GC?

 

[DIYer]

Hey DIYer I'll have to check them out. Did you post on IC or GC?

No its a private journal i only got on Google Documents at the moment. Well just the under the bed grow actually, i'll do the PC grow box journal later as its going to be pretty simple and need not nearly as much said about it as the whole under the bed rig is. I'll PM you a link

 

[SupraSPL]

This Qleaner bud caught my eye. It was flowered under a mostly dead HPS for the first half and finishing under LED. Small bud but it has some nice frosting coming together now

 

[SupraSPL]

Here are the rest. 4 different phenos of Qleaner, 1 Durban Poison, 1 Blackberry. Each is in a 1.5 liter pot.

 

[SupraSPL]

Still fattening up

 

[gates3rd]

Good job man !

Did you see that 100watt and 50 watt high power led`s on the bay ?
Are they good or better stick with Cree`s ?

The Cree MC-E M with 752 lumen are they good ? there arent many colors aviable but i think thats just a matter of time or ? My favourite would be "4 x Cree MCE4WT-A2-WG-M0 on round PCB" for my veg lamp. Maybe its a wrong direction i am heading hehe.

I really like your lamp and i like the fact its selfmade, realy cool, i wish i had made one aswell. Ill watch you and your lamp and hope to learn more from you to build one myself oneday cause cant get any nice LED lamps here anyway.

Must feel good to grow plants under a light you made yourself and you can change and extend whenever you want, till you found your own spectrum. You even can upgrade when there are new generation of led`s, while the premade LED still come with yesterday technology.

 

[SupraSPL]

Thanks Gates! I have heard of the high powered LEDs but I have not checked out their specs thoroughly. One problem that comes to mind immediately is heat transfer. We measure the rise in junction temp based on the resistance of the thermal path and multiply that by the dissipation wattage. That will give a huge efficiency advantage to small LEDs dissipating ~1-2 watts like most of the golden dragons we are using.

Also I have found that the 1-3 watt is the perfect light density to give a good compromise on yield/trimming and electrical efficiency because we can spread the light across the canopy very evenly. A higher wattage emitter would suffer losses if a reflector was needed. This would probably erase much of the advantage that LEDs have over HID. We need every scrap of efficiency we can get to compete with the big ones.

Cutter has the Cree MCE with a nice range of tints available for about $13 each. The cool whites they have are 430 lumens min and the neutral 370 lumens. From the spec sheet it looks like they dissipate about 5 watts when running at 350mA so that is about 80 lumens per watt. On the other hand the single die XP-G R5 is offering 132 lumens per watt at 350mA but cost about $8 a piece so they end up being a more expensive but much more efficient option than the MCE package. KNNA has some really nice cool white golden dragons right now that are cheap and easy mounting for DIY. You can mix them with some reds and make a vegging lamp or mix them with even more red and make a flowering lamp. Check out his LED thread on gardenscure if you have not seen it.

Yes it is fun to grow with a DIY lamp. The thing I like the best is being able to get good gr/watt numbers on a small scale without needing a 600w HPS with extractor and scrubber etc. LED really is great for micro growers. I still need to work out some soil/water/nutrient issues before I will get to see the full potential of the lamps.

With lots of help from KNNA I am working on some new lamps and will be testing them soon. I will try and retrofit this lamp to keep it up to date with the new LEDs coming out.

 

[blimblom]

SUPRASPL you have certainly mastered the LED DIY part of the equation. If you master at the same level your gardening skills, I think you will become a SUPRAgrower, and of course silence the commercial pest spinning everything out of proportions.

So my congratulations, I ll try my best to match your craftmanship


1.As I understand you used 35W /sqfoot which is aprx 32x32cm. Am I right?
(btw nice setup with the copper tape so neat)

2.Are you using only top light source? Im planning Scroging so probably light from the top will be ok.

3. As I understand you used 9:2:1 ratio for R/W/B . I read from someone else that 6:2:1 was his optimal for flowering. Can you spare your 2cents on this

4. about the driver I got a little bit confused. you used a driver of 120/48V 700mA on your setup?? You connected the LEDs in series till you reached the 48V? (or less just for being safe?)

5. your heat sinks are awesome, bit expensive though, I'll try to save some on that front (and be less elegant of course)

I highly respect KANNAs efforts , and I will probably ask him about supplying me too. Is it best that I reach him through gardencure?

thanks again for all the info shared and keep up the good work

 

[knna]

Hi blimblom,

I passed by here so I think I can aswer some of your questions related to the pure "technical" side.

3. As I understand you used 9:2:1 ratio for R/W/B . I read from someone else that 6:2:1 was his optimal for flowering. Can you spare your 2cents on this

Usually what you want to get is a given spectral distribution. Depending of the LEDs used and how you power them, you need to use a different number of each to achieve that given spectral distribution.

The higher number of red LEDs in this case is due they run softer than B/W ones. If you run them together on same circuit all running at same current, then you need to use less reds to achieve the same spectral distribution.

4. about the driver I got a little bit confused. you used a driver of 120/48V 700mA on your setup?? You connected the LEDs in series till you reached the 48V? (or less just for being safe?)

Drivers SUPRASPL is using are rated to 20-48V/100-700mA, constant current (adjustable). When you use a constant current driver, you can mix different LEDs on a string, as far as they can run at the same current. The limit of how many LEDs you can put is the max voltage output of the driver.

But these drivers are a little special, due they not limit the voltage output at all (technical explanation is complex, related to circuit topology), as is usual on most drivers. As far as the voltage load (added Vf of all LEDs on the string) is below mains (120 or 220VAC) rectified voltage, they work fine (indeed, they work more efficient as higher the voltage load). Of course, it has some disadvantages, mainly that max current output wont reach 700mA when voltage load is over 48V.

But most LEDs dont like to run so hard, and those which do it fine, anyway works better at lower currents. Red ones, in general, not do well over 600mA. In general I advice to run them at max of 570mA or so, and if possible, a little softer. These drivers are generally capable of giving those 570mA with loads still over 100V (although there are some that accept just 80V for that; its the problem of running something out of specs).

5. your heat sinks are awesome, bit expensive though, I'll try to save some on that front (and be less elegant of course)

Yep, he is using awesome heatsinks, and that guarantees excelent operation of the lamp. Always that is possible, it is better to use oversized heatsinks. In general, I tend to do recomendations of heatsinking a little oversized.

Price difference to use less heatsinking is not large at all.

What is possible to lower heatsink cost is to use other things that pure "heatsinks". Aluminiun profiles bars for windows are widely used for this purpose and works perfectly. Normalized aluminiun profiles (U, doble U and squared and rectangular tubes) are easy to get and generally they are way cheaper (on a sq inches of surface basis) that pure heatsinks.

Most people that did their own lamps using my components used one of this alternative options. On the other hand, when budget is not a big issue, pure heatsinks allows you to make very solid and nice lamps.

(note: if you have a low post count and cant send me a PM here, look for me at greenpinelane or the GardensCure or cannabiscafe.net (knnabinoide there). Anyway, after many time promising it, Im going to finaly open an online shop of components, I hope very soon)

 

[guest456mpy]

SupraSPL and knna,

First SupraSPL,

What to say about such a beautiful build.

Your heat sink is massive, but you have no fans, kudos!

knna,

I am pushing 500 watts of DIY in a 1m x 1m x 2m tent.
Previous grows were with a 300 watt unit in the center of the tent, I found falloff around the edges limited the way flowers were produced and matured around the edges. Harvest was good with tight inter-node spacing in veg and tight budding in floral.

I built two smaller 100 watt units to augment the central light. This is my first grow cycle with them so no data as of yet.

Anyone just reading this thread, we have nothing to gain, personally. LED's do indeed work if you are actually using proper hardware. It's still the "wild west" out there so don't just buy the first thing you see on e-Bay. LED-Girl did have a product that worked, but to say that it was on the bleeding edge of technology would be a misrepresentation of reality even 1 year ago.

Go forward in careful steps and you will be rewarded!

 

[blimblom]

oh thanks for all the enlightening info Knna.

as for heatsink, I'll definitely go some cheap alternative,

and your reasoning for not driving the LEDs "hard" (like 700mA), is the shorter lifespan, or the worse lm/watt ratio?? . Because if its lifetime we are after, I think its not worth it. As I think you said in another post, we are waiting for manyfold increase in efficiency in the next 2-3 years. Which means that it is probable that will scrap the old ones before they are through their usable life.

Even 15Kh (half their usual lifespan), is more than 3 years of daily 12/12 use. As a recreational grower, i ll never reach that figure.

I hope you finally open the online shop lol, I'll certainly need your products in the near future.
thanks again for sharing


PS Hempguy you are right in your word of caution. Lets not get things out of their proportions. LEDs are cool (literally) but they are good only for small grows yet. And also are considered way to "cool" (metaphorically) to be properly priced.
I will have a new small grow room, so after VerdantGreen new adventures, I understood that for my needs a 150W HPS is less cost effective than a DIY LED array.

If I already had an HPS system, I would wait for a year or so to see how things are going in the LED arena. definately they wont be worse.

 

[guest456mpy]

and your reasoning for not driving the LEDs "hard" (like 700mA), is the shorter lifespan, or the worse lm/watt ratio?? . Because if its lifetime we are after, I think its not worth it. As I think you said in another post, we are waiting for manyfold increase in efficiency in the next 2-3 years. Which means that it is probable that will scrap the old ones before they are through their usable life.

knna speaks the truth!

http://en.wikipedia.org/wiki/Haitz%27s_Law

That is, in part, why I don't push LED's yet. We, the pioneers, are literally paying the way for the future. While it doesn't make financial sense yet, it soon will.

 

[knna]

Yeah, exactly, running reds softer is related to the drop in efficiency, not to the useful life. Still running hard, good brands LEDs are able to have useful lifes over 35Kh, way more than probably any of us will use them.

Most red LEDs I tested, experience a noticiable droop in efficiency at some point between 550-600mA. A few a little harder, but always below 700mA. Thus running red LEDs, still if they are rated to work at 700mA or 1A, as those I usually use (SupraSPL LEDs are all rated to 1A continuos operation), at 700mA dont worth. Its a problem because most comercial drivers has two avalaible outputs, 350 and 700mA. At 350mA, LEDs works great, but you need a lot so it strongly increases cost. At 700mA, efficiency has dropped too much.

Ive measured some red LEDs an absolute output at 700mA below 5% higher than at 600mA, while power burned is about 18% more. That is why I recomend to run them below 600mA.

If you are going to use a fixed constant current driver to run "3W" red LEDs, I would advice to use a 500mA one, its a very good compromise between total light output for each LED (cost) and efficiency

 

[knna]

Hempyguy, I very glad to see other people able to do their own DIY lamps

Do you have a journal or an album where I can see it? I would say you are using as much power as I consider allowed to use on a given space. Of course, it depends of the efficiency of LEDs used, but Im sure that you are using good LEDs (because you look as an informed grower )

I cant agree more with you about that we are mostly working for the future. Currently, only on small spaces (PC grows, small setups) LEDs are clearly the best option. But they are so different on some things that we need to find how to get the best of them. Grower using HIDs lasted some years on finding the optimal amount of light to use, adapt feeding to the new spectrums, etc. We need to do the same with LEDs, but unfortunatelly people smelling cash has contaminated the research on the topic along the MJ forums. LEDGirl is not one of the worse vendors about it. She have some very questionable marketing practices, but past an inicial moment of many wrong information, lately I notice she is improving on that (for example, reccomending right coverages for her lamps).

But the fact is those very imperfect units and the marketing and false "information" associated with them are hurting severely our communitary research about growing with LEDs. People usually dont know the spectrum they are using and never the total amount of light delivered. And without knowing that, it is very difficult to take any valid conclusions from different setups.

Couple that with the often "participations" of vendors (with known nick or many others) on threads where people is joining to share their research to bias the course of the thread just to push towards their products, and you get the current scenario where many people uses LEDs but there is no progress on the understanding of how to get the best of them.

And I stop highjacking Supra's thread. Its a nice journal and if we want to talk about general things, I think we should open an specific thread.

 

[knna]

Oh, and about the Heitz "law".

It is the correspondence for LEDs of the Moore law that described situation of computers evolution. After many years of computers continously decreasing price for a given configuration or increased perfomance for the same price as time passas, most people is aware that when you buy a computer, you are paying a plus for having it now. Because if you wait, you can get it later by cheaper.

On this time of maturity of computers, except if you work on animation, multimedia, CAD or so, any computer of the market has enough perfomance to do the normal task people needs. But ten years ago, it wasnt the same. People who need to do something minimally obliguing had to buy the latest computer, because it was the only able to do it.

With LEDs, we are now in those starting moments. Comparable with early 80's, where firsts personal computers appeared. Do you remember the Apple was initially a project of hobbyists? They ordered 200 main boards because that people joined to the project, wanting to build their own DIY personal computer.

I got my first computer on 1982. It was an Apple II with 64KB of RAM. There wasnt standarized operative systems, mouses were waiting on the distant future. Data were stored on bulky flexible disks, 64 and 128KB each, or in music tapes.

Comercial LED lamps currently are kind of that. Nothing to do with what we have in the future. With LEDs, it is required a minimal perfomance, because there are alternatives (computers not). Until exactly this moment, the goal of LED's manufacturers has been to reach perfomance levels enough to easily beats other lighting technologies. Its been accomplished during this year (although many products uses older LEDs, way cheaper).

From now on, we are going to have access to LEDs with very good efficiencies. Continuing with the analogy, personal computers using the 8086. This was the point when computers started to be present on all homes, as prices were dropping. When they dropped enough, almost all people has one.

Probably LED lamps will follow the same trend. Now that the minimal perfomance has been achieved, we are going to notice the price drop. From this point, as soon as LED lamps using the modern high efficient LEDs reach the market, each grower must decide the moment/price at he enters the wave. But sooner or later, all will have LED lamps.

(Note: fortunatelly to computers, at the beggining manufacturers havent got a competence of computers of half perfomance by cheaper as happen with LEDs. They were expensive, but all were good, and first costumers were happy with the investment because the knew what they were buying)

Sorry for the rambling, I accelerated as I was writing