What's new
  • Happy Birthday ICMag! Been 20 years since Gypsy Nirvana created the forum! We are celebrating with a 4/20 Giveaway and by launching a new Patreon tier called "420club". You can read more here.
  • Important notice: ICMag's T.O.U. has been updated. Please review it here. For your convenience, it is also available in the main forum menu, under 'Quick Links"!

Buying chinese LEDs for DIY panel: worth or not?

K

KRANKENWAGEN

Member
Hello all

Im in a process of planning a stealth LED box for growing indoor during winter. It'd be a server case, large but low, and will require some extreme training.
Now, with case being very low, LEDs with their (relatively) low temperature seams like a best solution.

BUT. I don't have the money to buy high quality cree or philips or even osram LEDs just for an experiment.

So i was thinking of getting chinese LEDs off of ebay. Since grand master knna suggests 30W per ft squared, i was thinking of upping that to 40w or 50w, to compensate for the lower diode quality.

So i ask you: do you think it's worth?
Should i invest into relatively heap-ish chinese LEDs with any exectations to get some decent yield? Note that i'm doing this for my personal consumption, and while i love getting massive yields, it's not my primary concern. I just love growing shit...

thank you all
 
N

Neekz

Member
Well what's the footprint? I have myself, and shall again very soon. Build quality is exact, and diode's the same as most proven reseller's. U can go through Aliexpress so you can sleep easy, and make sure you find the most transparent seller's (Ie tech spec's like driver current, color ratio's, len's type, cooling etc...) . Look for proven designs, and scrutinize them well. You can find HGL's manufacturer if you look right. Let me know the layout your thinking and ill look into some options for you.
 
K

KRANKENWAGEN

Member
sweet, thanks mate.

First, know that i'm technically rather retarded, i LOVE that shit but i'm fully self taught so i might be lacking in some relatively basic stuff.
That being said:

The box would be somewhere between 2 and 2.5 feet squared (depending on the internal layout of the box). So 90w-100w sounds like a healthy number, light wise.
If the experiment ends in a success, i can always switch some LEDs for igh quality ones.
I'd build panels myself.

I've n+been checking ebay furiously for the past week and... wel... they all look the same to me. There are even some suspiciously low priced osrams and crees there, and the numbers look good (right color temperture), but i'm really the one to rate them...

thank you for your time
 
N

Neekz

Member
Id go for more wattage if you want bigger bud's. At 42w in 1ft/2 the bud's I am getting are CFL like in size, like a fat marker. I have a 180w panel covering 2'x3', and the center 2'x2' is growing buds on par with the size I am getting outdoor, but way frostier. The bud's in the side lighting are about the size of my smaller panel as well, so it can stretch the footprint a bit if you want. I purchased one exactly like this with the spectrum tweaked a little for $220 shipped. I plan on using it in a 2'x2' tent eventually, or getting another for complete coverage. I am getting very vigorous growth with this panel and spectrum, just haven't been posting it till I can get a fresh run from clones.
 
N

Neekz

Member
Oh ish I didn't even red the title fully, DIY I can help, but prices are too steep for me. Around $3-4/w depending in m calculations, even with china diodes, and only half my desired wattages.
 
N

Neekz

Member
I would make 4 individual modules cause large heatsink's may be hard to find for you. I would use these 30w semi-integrated panel's for each module. Then use something like this, preferably 120mmx120mm mated with the highest cfm 120mm fan your noise limit will allow (Always better to keep things extra cool...). For driver's you can use this from the led seller, or you can look for cheaper similar ones on ebay, that run closer to the 1000mA max of the diode's. This should get you in the ballpark of 100w. The spectrum is okay, and if you could get a large heatsink, say 40cm square, you could space them out nice and fill in the gaps with white's and maybe some UV and IR down the line. I would remotely wire the driver's for space. All in all a good system I thought of, but too pricey for me ATM. But you do get Cree's ;P !
 
K

KRANKENWAGEN

Member
yeah, prices are rather high, that's why i was asking is it worth getting into this at all, even with cheap LEDs.
Especially now when you gave me your numbers, that wattage would cost pretty penny

I guess 30W per ft squared is only for premium LEDs...
 
K

KRANKENWAGEN

Member
Neekz said:
I would make 4 individual modules cause large heatsink's may be hard to find for you. I would use these 30w semi-integrated panel's for each module. Then use something like this, preferably 120mmx120mm mated with the highest cfm 120mm fan your noise limit will allow (Always better to keep things extra cool...). For driver's you can use this from the led seller, or you can look for cheaper similar ones on ebay, that run closer to the 1000mA max of the diode's. This should get you in the ballpark of 100w. The spectrum is okay, and if you could get a large heatsink, say 40cm square, you could space them out nice and fill in the gaps with white's and maybe some UV and IR down the line. I would remotely wire the driver's for space. All in all a good system I thought of, but too pricey for me ATM. But you do get Cree's ;P !
Click to expand...
uh, haven0t noticed the new reply, thanks mate!
 
K

KRANKENWAGEN

Member
damn, while these cree modules look tasty, they're indeed too expensive for me atm

if i'm gonna invest properly, i'm gonna invest into a 400W HPS (which i do plan, after the winter, since i live in rather cold part of world). LEDs are supposed to keep me occupied 'till spring...

BUT. One of those, + some cheap WW and CW leds spaced around could support a small plant...
 
N

Neekz

Member
I would skip the generic cob white's and go for some cree XML's for the 30w, and xp-e's for the neutral and cool white's. They putout more light for less power, and potentially less driver's to buy. If you can save a bit more I would go for a better china panel. The cost per watt is way more better. $200 ballpark will get you at least double the watts your looking for with good optics for dense bud's/better penetration. You can get 80-90w panel's and ufo's from $100-150. The main things you wanna check is that their running at least 600mA driver's, optical lens's if you want usable light past 12" from the panel, and a well rounded spectrum, red heavy. The cost's for a DIY will definitely add up, and if you don't drop the extra $ on the quality good's like Cree's and Meanwell driver's, you run just the same risk of driver failure or bunk chip's as the china panel's. You can still put that all together and still have penetration problem's if you don't use the proper optic's. For this design, the spectrum's there, the diode intensity is on of the best, optic's can be found, it is just too pricey per watt for most initial L.E.D. user's.
 
N

Neekz

Member
Here's a little timeline of my L.E.D. purchases.

1)I bought 5 of those 10w cob's, dimmer, and driver's from singapore for about $50. Right of the bat the drivers were crap and were DOA. They looked good, and quality from the outside, open it up and you see the rats nest of cheapness. Ended up wiring them to a 12VDC wall wart controlled via dimmer. First thing I notice, they are blazing hot at full power, hotter than CFL's. I dial em down to about 1/5th full power. They have grown my seedlings and clones well with little to no stretch. I quickly realized though that this setup would net nowhere near even .25g/w. I did research and came to the conclusion that the real benefit's of L.E.D. growth are mainly due to the narrowing of light to specific color wavelength's rather then try and mimic HPS or CFL. It sort of defeats the purpose of L.E.D. when one uses mainly white's to cover a broader spectrum, they do do a great job at filling in a bit for red and blue, and the absent green when one uses red's and blue's, but they are by no means 1:1 for red/blue specific wavelengths. So I move on to purchase 2...

2) The Blackstar 100w UV UFO, 50-2w chips, 6-band's, cover's 1'x1', sounds perfect. At .6-1g/w she sounded amazing (3 zips in 1ft/2? Nice! Not...). Threw my clones under there and a couple week's later I realized that this is not gonna get me more then 1.5 zips with dialed condition's. Excellent growth, good coverage, plenty of vigor, it fit the bill of L.E.D. promises. It just didn't have that oomph to plump the bud's, and only had about 12" of usable light from the panel. The bud's with this type of panel (300mA, non-optic.) grow true to form, about the size of CFL nug's (Think fat marker size...) with plenty of frost and nice and dense (Strain dependent of course.). I got another at a deal for veg, but still yearned for more. Then I got the green-light for a bigger project, now we get to the china panel.

3) This is the panel I recommended up above. 170-185w, 90* optic's, 600mA drivers, 11-band's, all the ingredient's to the recipe for fat nug's, all the way down the plant. I cant make a sure fire conclusion yet on it till I do a run with clones, but I am currently finishing some outdoor in it. All new strains so I have no idea how long they will go or growth characteristic's but what has developed so far in 2.5 week's is impressive. The 2'x2' directly under it had 3 1gal single cola plant's, put under the panel at around week 5 from outside. They had been on a massive pistil explosion ever since, not to mention the frost density is way thicker. I pulled 2 of em early like week 7, Indy pheno got 9g top cola, sat pheno 7g, plenty of sugar trim and popcorn, and they probably could have gona a few weeks more and doubled in size, I am just anxious to re-veg these ladies ASAP as they are nice and vigorous producer's. Off to the sides in the foot of space on either side there is a 4ft tall bush with like 10 head's, the cola's on this guy looks a tad bit bigger than the ones under the non-optic, 300mA blackstar. The bud's run all the way about 20-24" down the branches, with the panel 12" from the top's, not to mention this plant is mostly in side lighting. On the other side is a bed of clones that the trich's are ready but just keep packing on pistil's. I think alot of yield and growth factor's are strain dependent. Alot of negative's I am noticing on the plant's are phenotypical, and I imagine if I had a good quality cut, this light would do a QP no sweat. Ill keep you posted on it, and let me know if you want any pic's in action.
 
N

Neekz

Member
I am telling you guy's, cut out the ebay middlemen and go to aliexpress and deal directly with the manufacturer. You can hustle em down if your into that, and get all info youd need.
 
K

KRANKENWAGEN

Member
ok, thank you all for the info, it has been quite an eye opener

My aim was to build an under 50$ LED panel for a midi or large computer case, but it seams that it still can't be done, yet.
LEDs work, but cost pretty penny. :(

thanks again!
 
chileno1978

chileno1978

Member
Neekz said:
I am telling you guy's, cut out the ebay middlemen and go to aliexpress and deal directly with the manufacturer. You can hustle em down if your into that, and get all info youd need.
Click to expand...
Sure it works sometimes but not always, actually I do that and sometimes is cheaper in ebay other cheaper on aliexpress, others the vendor is the same.
 
chileno1978

chileno1978

Member
ahaha just chek on ali, is arround US$28RB and US$31WB, I pay 30 each (almost the same), now is a bit expensive on ebay.
 
N

Neekz

Member
Look at tebos brother in the micro! 3 chip's, 9w's, in a small pc sized garbage bin. I cant imagin it costing more than $50. And chileno, it will work all the time, Aliexpress does an escrow service, so they never touch your money until you confirm it arrives in the condition you want it. Any discrepancies and you can keep the item with a 50% refund. Better protection than ebay IMO. Yes it is a hit or miss if YOU do not look good enough and read between the lines. I am offering to check over anything ya'll want out and give you an opinion. L.E.D.'s are a science, and ask any scientist or anybody scientifically inclined, one little variable off just a hair could throw off the whole project/goal. It is quite sad seeing all these people get frustrated because they did not read enough and invested into something they don't fully understand. Like buying a Miata for drag racing, or a Mustang for gymkhana, don't blame the vehicle for not doing the job you wanted it too. Not to say either of those can't do the other, but both car's would require some extensive modifications to hang on each other's home field. Car forum metaphor on the grow forum, might not be a prime example lol. Keep paying the drop-shipper's, or the ad-hype guys on eBay. Shit I may jump on that bandwagon and make a quick buck. Too bad all the nay-sayer's are out there crying wolf giving L.E.D. a bad name because they ordered something cheap and expect high-end result's. Again guys, no negative vibes, just would hate to see another person dissuaded by making a wrong purchase for their need's.
 
You must log in or register to reply here.

Latest posts

Latest posts

Top