Hi all,

I've been thinking about growing with LED's for about 6 months now and finally got around to building the board. I had initially grown a couple of times in this box with compact fluorescents but

had virtually no ventilation the first time (=mouldy bud), then scorched my plant the second . you can see from the pictures that for the 3rd grow i had guards around the lamps to solve this.

i also have a 12cm pc fan running intake and one running exhaust, which provide very good airflow. the crappy diagrams give an idea of the layout. arrows mark airflow direction.

the 3rd grow yielded 10.5g dry bud (i didn't put in the guards til i lost about 6 bud sites to heat so would have been a bit more). strain was Greenhouse seeds Hawaiian Snow (fem). smoked

well. given the volume of space used, in a cubic metre the equivalent yield would have been about 370g.

here is the spec for the LED grow:

strain: Greenhouse seeds - Arjans Ultra Haze #2 (i know its not same strain and will make comparison more difficult, but i gave the rest of my Hawaiians to my mate!)
grow volume: 35cm(w)*35cm(d)*35cm(h)
ventilation: 2* 12vDC 12cm PC fans
container: 8L Tupperware ~30cm(w)*30cm(d)*9cm(h)
grow medium: 3L clay pebble base, 4L house plant soil (both B&Q sourced)
nutrients: biobizz grow and biobizz bloom
light: custom LED board

LED board components (all sourced from http://rswww.com):
1* LED,3W,Emitting,Royal Blue,LXHL-PR09 (Luxeon, high dome) (450nm peak)
5* Blue,super flux,led,6000mcd,PCB mount,470nm,30deg,L-7677C2PBC-H (467nm peak)
40* LED,indicator,5mm,water clear,ultra bright red,L-53SRC-E (660nm peak)
10* Orange,super flux,led,18000mcd,PCB mount,630nm,40deg,L-7678C2SEC-H (640nm peak)
1* PCB,test board,copper clad strips,100x220mm,CIFAJB22
5* IC,Voltage Regulator,3 terminal,1A,positive,L7805ACV
5* Capacitor,ceramic,multi layer,axial,conformally coated,Z5U,20%,50Vdc,0.33uF
5* Capacitor,ceramic,multi layer,radial,2.5mm pitch,conformally coated,Z5U,20%,50Vdc,0.1uF
40* Resistor,metal film,0.125W,1%,91R
10* Resistor,metal film,0.25W,1%,51R
5* Resistor,metal film,0.25W,1%,100R
1* Resistor,wirewound,vitreous enamel,6W,5R1

the 40*660nm red leds are arranged in a 5*8 grid across the board. one half of the board contains the royal blue 3w luxeon in the center (of the half). the other half contains the 5*467nm blue leds in a die-five pattern (like on dice). each half also has 5 of the 10*640nm red arranged in a die-five pattern. the purpose of this arrangement was to see if any localised difference in plant growth could be observed under the influence of each half of the board i.e. to find the more cost-effective blue light (both combinations cost ~the same).

i am using a 400w PC PSU the power the LED board. A 12vDC rail is connected to the 5v voltage regulators though i honestly do not think that the regulators are needed as the pc power supply seems to be well regulated itself (my current one sits at 5.10+/-0.02V according to system monitor). i just put them in just in case. here is a link to use the PSU as a power source - http://www.wikihow.com/Convert-a-Computer-...ab-Power-Supply (they can put out masses of current, are cheap and reliable). the voltage regulators are heat sunk and cooled by the ventilation. this will also help the performance of the leds. in this case it is all just a bodge though

leds are wired in parallel as i didn't have the electrical confidence to set up the voltage regulators as current limiters. maybe next time.

15cm seems to be the best height for the led board above the plant as this is where the light first becomes evenly spread. i dont think i will be able to keep the plant under this level though as i only have 26cm for plant height(-led height so really about 21cm), but we will see. i want to compare against the previous grow in this box as that is all the LED board will effectively cover, so will only supplement in my bigger box.

anyways just as no one has really put up much info about led grows i thought i would add my 2 cents. any help at all will, of course, be greatly appreciated.

time for some chillin on that Hawaiian yum yum



Pete

grow box diagrams and compact fluorescent plant and bud.

led board and box pics.

the seed went into the soil on Fri 7 Sept. seedling sprouted on Tue 11 Sept.

Interesting setup, good luck with your grow

cool , i cant wait to see how it works out. i love small grows like this!

will be nice to see how these LEDs perform in a real grow setup

good luck

Gotta love a micro experiment.... but mate, the strain you've got - Arjans' Haze 100% sativa, 14-18 weeks to finish and in all probability will smoke gash - just not worth it bud think again.

B

cheers for the feedback everyone.

ah right, didn't know that about the strain. oh well. this is turning out to be comical already!

it is just an experiment after all though. i guess i'll see what happens and post back if it wont make me look more foolish

thanks all,


Pete

if there are leds involved, you have my attention. my chair is always here for led grows.

If you don't mind me saying, that isn't a lot of light for a square foot ....

If you believe LEDs are at least equivalent to fluorescents, you ought to be aiming for Blab's 80 watts.

... though I used to grow at very low light levels when 25 to 30 watts PSF was the norm.

40 ultrabright reds - thats only a couple of watts, surely ?

If you look at the lettuce growing example I keep posting, you need a lot of cheap and cheerful reds - more like 400 to grow a complex, small-leaved plant.

Click to view attachment

http://www.metaefficient.com/archives/hydr...ow-lettuce.html

As for all this precise spectrum matching .... maybe I would think differently if I did a MH grow, but I get such good results with my over-red sodiums, I can't see myself in vesting unless one of my ballasts packs up ..... though I've considered scaling down ....

I still can't see the appeal of using lights patently better suited to growing algae than plants, (and I bet I could grow more algae with a cheap fluorescent for the same electricity ....)

.... but maybe I'm just old and jaded....

.

Hi Hackmagic,

Wishing you all the best with your grow. I will defiantely be pulling up a chair on this one, so keep us posted how it goes.

Have u looked at the lumen output of your rig?
mcd to Lumen convertor

One of the red leds (of which you have 40) is putting out 0.75 lumens.

So in terms of total lumen output, all of the leds together are prolly putting out less than 100 lumens. However, because you have hand picked the light spectrum, i bet you dont need as much for a successful grow.

Looking forward to seeing your results.

BAB

Thank you buildAbong, compostverte and LED_Experiments, i was gonna snip and start from scratch after being laughed at (the shame ) but i shall continue (not that there is much hope )

compostverte, i was pretty sure that it wasn't going to be enough but i wasn't prepared to fork out loads just yet. i'm quite satisfied i can scale up and make a much better board now. your over-red sodiums prob are the best balance between cost, ease of use and efficiency available today (just having a guess!), but i shall look to the future (which is hopefully bright, well-balanced and mostly red with a bit of blue ) plus i like experimenting.

thanks for the link buildAbong, it will be useful for the future. most of the LED specs quote total output as light has been measured using an integrating sphere (whatever that is) and give a relative distribution. plus the leds are 15-20 degree (either side of centre, with light at half intensity) so i estimated that useful light is only output over about 1 steridian, so i was going with the straight values! this is probably stupid but it was just a guess. at least i can calculate the output a bit better now.

although embarrassing i will post a pic of my feeble (it really is) plant. it is growing but the stem has been weakened by stretch as prob only 1 or 2 leds worth of light was getting to it initially. i added some makeshift reflectors and it has stopped, and it is producing leaves, albeit slowly. i do expect better performance as the plant grows and can use the light more effectively. for now i will just hope my little dude thickens up some time!

Thanks all,


Pete

Hi,

I noticed that Osram released the Ostar led package that can approach 1k lumens with the right reflector. The are about £30 each and i cant find a price for the refelctors, so i guess they are expensive.
Ostar datasheet

You could literally place your 100 lumens right on the plant and see what happens. It would be worth a try

Light intensity is proportional to the sqaure of distance, so 100mm from the plant @ 1k lumens is the same as 100 lumens 10mm away (or am i stoned and not thinking straight?)

BAB

Hey buildAbong thanks for the link. they seem pretty powerful leds but they are white so spectrum contains a lot of green. farnell only seem to have white but i looked on osrams site and there are red (617nm) and blue (464nm) projection leds which look pretty good. they recommend you heatsink and fan each one (3cm*3cm heatsink, 2.5cm*2.5cm fan - see http://catalog.osram-os.com/catalogue/cata...02d52c000100b6) which is do-able but a bit of a challenge for me at this stage!

on the thermal management note the Luxeon bulb has packed in. i guess it didn't like being driven without a heatsink for 24hrs a day! another lesson learned. tum-te-tum...

only just read the "All about LEDs" (http://www.uk420.com/boards/index.php?showtopic=79544) thread there, its very informative. it didnt come up when i searched uk420 for "LED". oh well again it will help for the future


Pete

looking cool mate love to see the result cant wait

Hi again,

Have you managed to find a retailer for the coloured ostars?

BAB

Hi buildAbong, sorry i haven't been looking this past while, got a new bike last week been checking it over rabidly! given how my luxeon failed i really think that heatsink as recommended for the ostars would be necessary. i'd still be quite keen to work with high power LEDs but i haven't really thought about the next version of the board except for active cooling possibilities and ideas for requirements for lighting levels and led densities.

just thought i'd post some new pics too of the little dudes progress. first one is 19th Sept.

Pete

2 from 21st Sept.

2 from today, 25th Sept. the plant is 2 weeks old here.

Hey all,

just thought i'd let you know how things turned out. a rather early finish eh? 3 weeks of flowering and no changes except a freezing of growth. at this point i chopped the plant.

i'd like to ask a few questions about the flowering.

with my previous miniature compact fluorescent grow it took about 3.5 weeks for flowering changes to occur. the plant then was better developed with better stem growth at the fan leaf nodes. i also introduced a 75w Grolux incandescent bulb at about 2w 5d and then flowering occurred about 5 days later. i was wondering if the plant needed time to build up hormones to initiate the change and it took longer than as it was small/young, or was the introduction of the bulb key? i assumed the bulb was key.

in this LED grow i introduced the 75w Grolux for the last 15min at the end of the 12h day (to supplement the far red i hoped, as i had presumed the bulb was key). would the underdevelopment of the plant extend the initial flowering period further still or would the bulb need to be on the full 12h day to trigger the flower? perhaps i should have left the grow run on.

thanks for any info on this. it will help me to control things better in the future.


ah well yet another failed led grow! i'll try again when i make a better light though costs of £100 per sq ft are quite high compared to hps!


Pete


pics below on last day of 18-6

another on last day of 18-6

pic after 3 weeks of 12-12

Well nobody will get a suntan in a coal cellar eh?

Need a lot more lumens dude.

about your questions
for better growth at fan leaf nodes (secondary branches) you should use LST and get your plant growing horizontally. in that case new branches get much more light and they develop much quicker. you also need to place your lamp as close to the plant as possible. even if leds touch the plant there is no problem.
flowering delay: when you switch to 12/12 it can take 2-3 weeks before your plant start to produce flowers. you can make that delay shorter if you use different day/night ratio, let's say 11.5/12.5 untill your plant starts showing sex. you can also use halogen lamp at the end of the day cycle to give the plant a bit of 735nm. ofcourse if you put a small plant to 12/12 it also needs some time to start flowering (i think), if you have better developed plant, it takes a bit less time.
your experiment didn't fail, you just made some mistakes and found one way things don't work. don't give up on leds, try another grow, learn by mistakes and experiences.

Looks like you need higher intensity. I recommend using high power led's, Compared to the wattage used it might be cheaper (to build a light) than 5mm, which have less light intensity coming from a single point. instead of halogen to get far-red light you can get far red leds. Keep working at it, sooner or later a technology will be developed thats better than HPS.

Just as Edision put it "invention is 1% inspiration and 99% perspiration" .