LED FAQ) Building and DIY

[Dion]

whoa 2x 3070s in a 9"x18"at 60w?!?! so thats 1.6A? for 15800L in just over a sq ft? seems pretty bright yo but i love it
unless...what binning is that?

@ blynx
i forgot to mention the cooler you can keep them the more light they give you



so unless your cooling in this cab is spot on you could consider running them at a lower current and end up with the same Lumen output(they become less hungry for volts when they r hot)

grouchies idea for putting the sinks into the ceiling of cab is good in a micro grow however i have run into opposite trouble where id flip the ventalation around( ie pull from the top and push out the bottom) to get temps warmer-will really depend on your unique situation

 

[neverforget4/20]

bro,

are you convinced it's wortwhile to use the reds? Im considering very similar idea for my diy.

 

[Dion]

bro,

are you convinced it's wortwhile to use the reds? Im considering very similar idea for my diy.

no,

plants like photons

refer to beta's work here ic forum

 

[grouchy]

They are AB bin cxa3070. I could probably benefit from running co2 with that much light.

 

[blynx]

I've read that not all dimmers work with LEDs. What dimmer do you guys recommend for use with LEDs?

Thinking about veg lighting. Do the 3070s come in 5000k?

 

[Dion]

I've read that not all dimmers work with LEDs. What dimmer do you guys recommend for use with LEDs?

Thinking about veg lighting. Do the 3070s come in 5000k?

dimming is expensive-so i cant really say- peeps spend $40-70 on drivers i spend 8 so... perhaps someone else will chime in id advise to just run them lower from the start

as far as not all working- u want a dimmer that adjusts the current not volts

yes the cxa 3070 is available in 5000k

BUT if you want a veg light
id go to fasttech.com and get some xmls or even the 5 pack of xp-gs they have

or get smth chepaer than a cxa3070
like a xca 2540/2530/2520(all available in 6500k)

 

[grouchy]

The xm-l2 would be a more efficient chip than the xml or xpg but may be a little more expensive. That's if you were wanting to run smaller chips. I think a 3070 in 5000k or 4000k would be a great veg light personally. I use a dimmable meanwell driver for my cobs that goes up to 1950ma. Part # is HLN-80H-42B with a 100k potentiometer for adjusting. It allows me to run them up to 80 watts each.

 

[Dion]

yes the xml 2 puts out an extra 12 L (depending on binning)
i suggested those 2 from that site because of price also running at low current (as we dont need too much light in veg at all) efficiency will be awesome anyway

also fastech just got in a cxa cob for like 10 bucks no binning though

why do u think the 3070 would be a great veg light?what are its benifits in this situation?

how much was your driver btw?

 

[neverforget4/20]

The xm-l2 would be a more efficient chip than the xml or xpg but may be a little more expensive. That's if you were wanting to run smaller chips. I think a 3070 in 5000k or 4000k would be a great veg light personally. I use a dimmable meanwell driver for my cobs that goes up to 1950ma. Part # is HLN-80H-42B with a 100k potentiometer for adjusting. It allows me to run them up to 80 watts each.

grouch is tbat 1driver per cob? Does the dimming function generate heat or loss of efficiency?

V2 vero 29s are now available.

thanks for sharing

 

[Dion]

https://www.fasttech.com/products/1612/10004218/1320606

good for a cxa 3070- i still favour the one in the intro posts though especially as the seller will dim it to a lower current


here is an afordable dimmable driver( it goes up to 3 A though so i would prob wire it in paralel with 2 cobs- if ur like me its only a matter of time before im stoned and think"why havnt i cranked thos baby up? lets see how bright she goes"..... sizzle pop....."oh thats right")
http://www.satisled.com/70100w-adju...input-dc3136v3a-output-with-dimmer_p1542.html

blinx id consider this with 2 cobs for your veg light

also the vero 10s seem to be value but the larger ones ie 29/18 not so great

 

[Dion]

https://www.icmag.com/ic/showpost.php?p=6893898&postcount=98

there is a test done by the dopest on the drivers i linked in the intro- the ones i use fro china

http://www.aliexpress.com/item/Wate...-supply-lighting-transformer/32265565611.html

Input:
122.4 volts
.48 amps
59 watts
Power Factor .99-1.0

Output
1.481 amps
37.98 volts
56.263 watts

here is the whole thread lots of back n fourth about this build- perhaps useful to others

https://www.icmag.com/ic/showthread.php?t=300225

 

[OIBI]

The penetration just wont be as good as HPS
if you wanna grow trees you can't just have a huge LED spotlight to do this, you need multiple smaller lights closer to the tree. I would assume that to light more than 30cm of canopy thickness will result in light bleaching on top and fluff down below

This simply isn't the case with larger cobs. 'penetration' with the larger cobs when driven at nominal power levels do every bit as well. Especially if a lens or reflector is being used.

The heat is still there

on average LED lights are 20% more efficient than HPS lights that doesnt equal no heat, it equals less heat-meaning they convert energy( electricity) into light better. where as HPS could convert 100w into 20w of light and 80w of heat LEDs will convert 40w into light and 60w into heat BUT the thing is the heat wont be thrown off the light like with HPS it will dissipate off the back so you wont need to cool the space to as low a temp. think of it this way, in a room that is 28c under an hps light it could be 33c so you really have to keep that room cool to not cook your girls( in reality the room is probably more like 25c but we are measuring at the canopy-read under the light) but with LED in a room that is 28c it will only be 29 or 30 under the light. In reality you will have only a little less heat to deal with but it is cheaper because of the way you have to deal with it ie less watts for Aircon/venting because our room can be 28c and still have the same temp under the light. (it is probably important to note that because of this you should run your led rooms a few degrees warmer than you would with hps, this will also cause the girls to drink and eat more.many led growers report needing to use less nutrient with leds. get ur temps up and this should not be the case,it is possible that many sub-par results from led grows were contributed to by this temperature anomaly causing plants to eat and drink less therefore grow less)

'Heat' is a function of power displacement. A 600w HID is going to produce the same amount of heat as a 65% efficient LED or 600w worth of incandescent bulb. The advantage of a higher effeciency lamp is the amount of radiant energy the plant can utilize before ending up as thermal energy. Since ultimately most of the radiant energy absorbed by the plant is released as thermal energy in order to support metabolic functions.

In otherwords, you are correct. The 'heat' is still there, however replacing 600w of HID power with 600w of LED power is not the solution to that. In fact if you're using an air cooled reflector, it would actually make the problem worse. The trick is to design a lamp of an adequate efficiency so that you can lower the total power displacement, while maintaining the same same photon count at the canopy. So 300-400w of 45%+ efficient LED to replace 600w HID's would be a better approach.

Chips- I recommend cree Cobs, they are the best and although the price reflects that you get what you pay for, also cree COBs use a ceramic board making thermal management child's play you can buy them from digikey or mouser. I also recomend surexi chips by illumitex they are great and the company has invested enough in R&D to have a nice range of horticultural chips( http://www.illumitex.com/illumitex-leds/surexi-horticulture-leds/ ), these however are no longer in production as they now are just selling prefab lights but you can still buy them on amazon or mouser while stocks last. Others use vero( bridgelux) and you are welcome to try them I never have and never will. Cheap chines cobs off ebay or aliexpress? nah we are chasing efficiency here leave the junk in china.

Don't be so quick to dismiss the Bridgelux vero lineup. While the CXA/CXB series from cree has an luminous efficiency advantage when driven softly, bridgelux cobs actually have a superior thermal package. The aluminum substrate use by bridgelux is quite marvelous. When driven at nominal current levels or above, they tend to suffer from thermal droop than the comparable CXA/CXB package.

Not to mention ease of assembly with the pico EZ-mate and built in cob holder. Makes it very easy to use the kapton tape for those unable to drill and tap. The vero 18 is a fantastic chip for those new to building a DIY lamp. It's cheap enough that almost anyone can justify playing around with one. The vero 10 is compatible with any heatsink designed for 20mm led stars which can make retro fitting old blurple lights a breeze.

 

[OIBI]

Great write up otherwise. Thanks for posting this.

 

[Dion]

This simply isn't the case with larger cobs. 'penetration' with the larger cobs when driven at nominal power levels do every bit as well. Especially if a lens or reflector is being used.



'Heat' is a function of power displacement. A 600w HID is going to produce the same amount of heat as a 65% efficient LED or 600w worth of incandescent bulb. The advantage of a higher effeciency lamp is the amount of radiant energy the plant can utilize before ending up as thermal energy. Since ultimately most of the radiant energy absorbed by the plant is released as thermal energy in order to support metabolic functions.

In otherwords, you are correct. The 'heat' is still there, however replacing 600w of HID power with 600w of LED power is not the solution to that. In fact if you're using an air cooled reflector, it would actually make the problem worse. The trick is to design a lamp of an adequate efficiency so that you can lower the total power displacement, while maintaining the same same photon count at the canopy. So 300-400w of 45%+ efficient LED to replace 600w HID's would be a better approach.



Don't be so quick to dismiss the Bridgelux vero lineup. While the CXA/CXB series from cree has an luminous efficiency advantage when driven softly, bridgelux cobs actually have a superior thermal package. The aluminum substrate use by bridgelux is quite marvelous. When driven at nominal current levels or above, they tend to suffer from thermal droop than the comparable CXA/CXB package.

Not to mention ease of assembly with the pico EZ-mate and built in cob holder. Makes it very easy to use the kapton tape for those unable to drill and tap. The vero 18 is a fantastic chip for those new to building a DIY lamp. It's cheap enough that almost anyone can justify playing around with one. The vero 10 is compatible with any heatsink designed for 20mm led stars which can make retro fitting old blurple lights a breeze.

tnx alot for the feedback

re penetration i was referring to needing multiple cobs as apose to 1 mega cob( ie 6x 50w cobs not 1 600w cob)- ill go back edit and explain that better
would you agree with that or am i missing smth?

as far as heat goes im a bit unsure if i misunderstand you, we r saying the same thing or we disagree- my point there was that leds arnt magic cool lights that dont get hot- energy in= energy out, just that more of it will be light and less heat and that the heat will not be projected onto the canopny as much so less energy needed for cooling/ can let the room get warmer

i am not familiar with the vero so ill take ur word for it- efficiency is what i chase so cree works for me- cree also has chip holders( with or without electrical connection) and the ceramic board i refer to IS easier for heat management in my experience compared to alu boards


all in all i really do apreciate the feedback because there is way too much bullshit on this forums led section and i just want to make it simple for peeps who are looking into diy- as diy led is simply the best price available for grow lights currently imo

 

[OIBI]

tnx alot for the feedback

re penetration i was referring to needing multiple cobs as apose to 1 mega cob( ie 6x 50w cobs not 1 600w cob)- ill go back edit and explain that better
would you agree with that or am i missing smth?

How do you edit something on this board? I can't find the button? I noticed a few typo's in my post....

Well even at 50w cob's actually 'penetrate' quite nicely due to the high amount of green/yellow photons in the mix. Compared too a 100w HPS, a 50w cob is a much better perpetrator. 10, 100w cobs would likely yield better results than a single 1000w HID simply due to the additional pathways for the photons to travel.

That's the nice thing about cobs, you have the flexibility too either distribute the light and have a low hanging height, or you can run them at higher power with a reflector and mount them 24-36" above the canopy.

I just think it's an incorrect statement to say they don't 'penetrate' as well.

as far as heat goes im a bit unsure if i misunderstand you, we r saying the same thing or we disagree- my point there was that leds arnt magic cool lights that dont get hot- energy in= energy out, just that more of it will be light and less heat and that the heat will not be projected onto the canopny as much so less energy needed for cooling/ can let the room get warmer

Maybe we're saying the same thing? My point was that 600w of power displacement is going to produce 2046BTU of thermal energy within the space. Regardless of if is LED, HID, or a toaster oven.

i am not familiar with the vero so ill take ur word for it- efficiency is what i chase so cree works for me- cree also has chip holders( with or without electrical connection) and the ceramic board i refer to IS easier for heat management in my experience compared to alu boards

Various testing done by several members of the rollitup forum have confirmed that the vero series run better than the CXA/CXB series when driven at nominal current levels and above. Even when driven hard and hot they suffer very little temp droop. I'll have to go look for the specific numbers again, as I can't quite recall.

I know that CREE has cob holders available. They're not always easy to find. All would require the heatsink to be drilled and tapped in order to function. Bridglux is noob friendly since they don't use a binning system, and the cob holder is integrated. It simplifies the build process and is super inexpensive.

Don't get me wrong. I prefer the CXB 2530 and 72v 3590 emitters. However the vero series can hold it's own and much easier to find. Most people won't notice the difference to be frank

all in all i really do apreciate the feedback because there is way too much bullshit on this forums led section and i just want to make it simple for peeps who are looking into diy- as diy led is simply the best price available for grow lights currently imo

There's to much bull shit in the indoor garden scene period. So it's good to help clear the air.

You did the hard part

 

[who dat is]

Thanks for consolidating the info to one thread, subbed.

 

[Dion]

@OIBI
as far as penetration goes i see your point and do agree BUT
in an effort to make this write up "noob friendly" i suggested not using lenses or reflectors and chasing efficiency in the build so: woudl be running the cobs softer and then hanging them closer to the canopy- i guess its a lil dif aproach just trying to get nice even light spread on canopy
ok so the heat thing we agree- in fact what you are saying was my point- however not as eloquently or technically explained as you did

if you have any graphs/ would like to post some information on veros( like which ones to get how hard to drive them etc) it may be useful to others- i dont know much about that- and they rnt binned? huh i remember bueno testing his( binning them) when he was wiring in parallel strings- makes sense now

thanks

 

[J-Icky]

Ok I was looking at the 2530s and was wondering if it would be possible to rum 3 of these off of a single 100w driver. I want to build something that will run at roughly 200w(more likely 180w running 6 2530s at 30w). The only problem I have is I don't want to run the cost of the build up by using 6 different drivers. SO would it be possibeto buld 2 separate units of 3 2530s using only 1 driver for each unit for a total of 2 drivers.

I know you're big one the 3070s but I seriously want to use the 2530s for better coverage as I'm limited on room. Also if I could run multiple 2530s on a single driver it would be ALOT cheaper to get 180w of real power running the 2530s at 30w than buying 3 of the 3070s and running them at 60w. So the 3070s aren't even really an option no matter what anyone says. For the price of the 3070 setup I could go buy a higher wattage of the other style led already to go, or a 250w HPS or an insane CFL setup.

any and all help on this would be appreciated.

 

[Dion]

Ok I was looking at the 2530s and was wondering if it would be possible to rum 3 of these off of a single 100w driver. I want to build something that will run at roughly 200w(more likely 180w running 6 2530s at 30w). The only problem I have is I don't want to run the cost of the build up by using 6 different drivers. SO would it be possibeto buld 2 separate units of 3 2530s using only 1 driver for each unit for a total of 2 drivers.

I know you're big one the 3070s but I seriously want to use the 2530s for better coverage as I'm limited on room. Also if I could run multiple 2530s on a single driver it would be ALOT cheaper to get 180w of real power running the 2530s at 30w than buying 3 of the 3070s and running them at 60w. So the 3070s aren't even really an option no matter what anyone says. For the price of the 3070 setup I could go buy a higher wattage of the other style led already to go, or a 250w HPS or an insane CFL setup.

any and all help on this would be appreciated.

ok so it sounds like you are pretty keen on the 2530s

is it possible to run them off 1 driver- yes
you can wire them in series if you find a driver with enough volts at the desired current or
you can wire them in parallel if you find a driver with too high current and want to divide it
i suggest you reading the electrical desighn section at the start of this thread then you would understand this questions vagueness yourself- i wont go driver hunting for you sorry but if you have a driver you are considering and want to be sure it will do the job link it and ill scope the specs

you should consider what current you want to drive them at after refering to the cree pct( http://pct.cree.com/dt/index.html )
at 30w that is 6 chips each at a current of 800ish? shoudl be easy to find a driver that gives up to 40v at 2400ish mA and wire 3 chips in parelel( that will devide current so you will get 3x 800mA)



what binning are you looking at?
btw
i can totally understand you wanting to use smaller cobs however i doubt it will be cheaper that way BUT i wont try to sway you: a man always has his reasons

what size is this for?

 

[J-Icky]

Ok so first thanks for the quick reply. As for a specific driver I haven't really looked yet. Right now I'm trying to decide if I want to build a COB led, buy a mars hydro 300w old style or build a CFL setup using a bunch of 23w bulbs. So for the driver , if I understand it correctly I need one that runs at the minimum of the combined volts of the series of cobs, example would be if they are 12v each cob, to run 3 I would need a driver that can do 36v. Is this right? Do I also need to look for combined amps too?

From what I've read on the 3070s is that people say not to run them over 60w for best performance. Even if I ran them at 80w it would cost me close to $100 on just the cobs when I could get the 6 2530s for$50-60. SO for me the only way the 2530s would be more to build is if I had to buy a driver for each chip.

Right now I'm just trying to find out as much as I can and while I can do all the labor, as in once I have all the right pieces I'll have no problem getting it together. Its all the numbers and technical stuff that I don't understand and obviously need to figure out before I start doing anything. So any and all help would be appreciated.