Lighting COB LED INFORMATION check it out!!!

[Dion]

I understand, thanks.
Are that numbers for heatsinks real? Or you invent them? Because they seems too big...

those r true numbers dude, thats what you want to calculate heatsink capabilities

 

[KonopCh]

well its not liner

lets say you have an LED that is 50% efficent at 50w(cxb3070) that means you have 25w oƒ heat produced so you need a heatsink that can disapate 25w of heat... if u run passive(no fan) you need 110cm2 per heat watt if active you need 40cm2 per heat watt

so for this cob you would need 2750cm2 of surface area for passive or 1000cm2 for active cooling to maintain aproximately 50 Tj

now if we ran the current at say 2000mA the eff would drop to around 45% and now we are running it at like 75W so 55% of that becomes heat so 41.25 watts of heat so need a bigger heatsink

You sayin' I need more than 50x50cm passive heatsink for one COB? That must be some typo here?

 

[Dion]

You sayin' I need more than 50x50cm passive heatsink for one COB? That must be some typo here?

50x50? how does that translate into surface area?

it you have a heatsink that is 5x5 with 10 fins on it each 2.5cm tall, what you have is: {(2.5x2)x10+5 +5 x 5} so that is the surface area (2.5cm tall fins x2 because both sides x10 because you have 10 fins and +5 for the width of base and +5 for the width along the top(top of fins plus space between fins)= perimiter perimeter x lenth(in this case 5)= surface area


so your 5x5cm heatsink with 10 fins 2.5cm tall= 300cm2 of surface area or enough to cool 7.5 watts with a fan on it,

note that is 7.5watts of heat, not watts from the wall, so a cxb3070 at 1400mA is 50% efficent, meaning that only half the watts from the wall convert to heat, meaning you can run 15watts of that on this heatsink


im trying to get spreadsheets posted so you can all just pkug ur numbers in and not need to do the maths

 

[pop22]

I went with the 2530s for the same reason i could afford to build using them, or not build simple as that. Yea 3590s are better. And rolls Royce is better than...;will just about anybody...lol, doesn't mean we can afford one! At least with we have the same basic structure to upgrade from. The 2530s get us started while we save/earn to be aqble to upgrade our chips. So, yep, in the long run, we'll spend more. But then again, how much credit card interest will someone pay out on the 16 CXB 3590s they bought??....

and I'm seeing the same thing here as on other forums, if your not building the biggest and baddest, your doing it wrong....... well, for those who can afford to run that way, hell yeah? But how about some thought to those of us that simply CAN'T afford it, at least not all at once. We need better light NOW, not a year from now when we've saved enough. So while anyone offering advice, please keep in mind, we need a range of options that will work, not just the baddest with the mostest...lol! My best with cheap, Chinese lights is .7gpw. Every point above that is a 10% increase in production. So if I/we can hit 1+gpw with a diy, even if the total harvest is smaller, it means by adding more cobs later, or buying more powerful chips, it will get even better. Those of us on a budget yet still wanting to improve our grow, can do so incrementally.

sorry to hijack your thread. I just want to see people offering solutions that may be compromises, but obviously better than generic burple lights, as well as top end gear.


And to answer your question, you can use that driver but you'll need to adjust the current down to say 700-800ma. The 2530 will run up to 1100ma but at greatly reduced life span. I'll run mine at 700ma.

Im on a similar budget bro, in a 1x1m2. so stay tuned. And the best start for a guide is growmau5 on youtube his diy led beginners guide is invaluable.

@BigSm0 Yo! Iv'e been looking around and clearly the 3590 are top dawg but thats over budget, 2530 look great but what about higher and lower powered units than the 2530 are any of those worth a shot? I've already got and onl cp heatsink so its just driver and cobs and im set!

also i know roughly what driver i need/want but only small discrepencies in numbers or letters to do with w\a\v\ model so on and so forth..
im looking at MW HLG -240H-c1050A and on ebay their is one going cheap second hand that i want to pull the trigger on but it is a c1050(B) not A and i dont know the difference yet any help would be appreiciated guys!

 

[BigSm0]

dont have to explain a thing man. Your choice is great. I'm not 100% sure on the specs on them but if you think 10-20% less efficient over the total wattage say being 200 watts it would take you along time to pay the more expensive ones off in energy savings. Plus your still way ahead of the converted panel.

Growing for me is a hobby that went out of control. Fortunately the end product is enough to reimburse me and any future build costs. But the one I'm building are all 2530's. Luckily cob tech isn't as bad as regular electronics. Could you imagine being outdated every month or so?

 

[KonopCh]

50x50? how does that translate into surface area?

it you have a heatsink that is 5x5 with 10 fins on it each 2.5cm tall, what you have is: {(2.5x2)x10+5 +5 x 5} so that is the surface area (2.5cm tall fins x2 because both sides x10 because you have 10 fins and +5 for the width of base and +5 for the width along the top(top of fins plus space between fins)= perimiter perimeter x lenth(in this case 5)= surface area


so your 5x5cm heatsink with 10 fins 2.5cm tall= 300cm2 of surface area or enough to cool 7.5 watts with a fan on it,

note that is 7.5watts of heat, not watts from the wall, so a cxb3070 at 1400mA is 50% efficent, meaning that only half the watts from the wall convert to heat, meaning you can run 15watts of that on this heatsink


im trying to get spreadsheets posted so you can all just pkug ur numbers in and not need to do the maths

Aaaah, thank you!! I only calculate surface area, so that's why I get big numbers for heatsink. I didn't calculate fins, only plain square of aluminium. Now I understand this, thanks!
Nice, spreadsheets would be awesome.

 

[pop22]

yeah I see that, sorry!

I'm also looking at the Citizen CLU-58 1825 for future upgrades. Its got some great specs.

yeah its all photons

I have some cxa2540s at 1200mA=38% eff
and i have a cxa2530 at 600mA= about the same

but if u consider running a cxb3590 at 100w its 42% eff roughly so

it all works dude/dudette

btw the spreadhseet u posted is for the cxb2530 not the cxa-are you talking about the cxb or cxa2530?

 

[Dion]

I went with the 2530s for the same reason i could afford to build using them, or not build simple as that. Yea 3590s are better. And rolls Royce is better than...;will just about anybody...lol, doesn't mean we can afford one! At least with we have the same basic structure to upgrade from. The 2530s get us started while we save/earn to be aqble to upgrade our chips. So, yep, in the long run, we'll spend more. But then again, how much credit card interest will someone pay out on the 16 CXB 3590s they bought??....

and I'm seeing the same thing here as on other forums, if your not building the biggest and baddest, your doing it wrong....... well, for those who can afford to run that way, hell yeah? But how about some thought to those of us that simply CAN'T afford it, at least not all at once. We need better light NOW, not a year from now when we've saved enough. So while anyone offering advice, please keep in mind, we need a range of options that will work, not just the baddest with the mostest...lol! My best with cheap, Chinese lights is .7gpw. Every point above that is a 10% increase in production. So if I/we can hit 1+gpw with a diy, even if the total harvest is smaller, it means by adding more cobs later, or buying more powerful chips, it will get even better. Those of us on a budget yet still wanting to improve our grow, can do so incrementally.

sorry to hijack your thread. I just want to see people offering solutions that may be compromises, but obviously better than generic burple lights, as well as top end gear.


And to answer your question, you can use that driver but you'll need to adjust the current down to say 700-800ma. The 2530 will run up to 1100ma but at greatly reduced life span. I'll run mine at 700ma.

dude dont listen to thos small dick fuckers

if u have a cxa2530 and run it at 20w per cob you have the same efficency as a cxb3070 at 50w


par is par

 

[pop22]

I'll have to check but I think Growmau5 did a grow with the CXA 2530 light before he used it to veg and pull 1.8 gpw! So yea, I know what they are capable of!
If I can hit 1.0 I'll do a dance! that's a 30% improvement alone! And I have three bars with one built and another almost done.

Then, a new light is coming...........

 

[BigSm0]

I'll have to check but I think Growmau5 did a grow with the CXA 2530 light before he used it to veg and pull 1.8 gpw! So yea, I know what they are capable of!
If I can hit 1.0 I'll do a dance! that's a 30% improvement alone! And I have three bars with one built and another almost done.

Then, a new light is coming...........

You'll absolutely pull 1gpw. I think supplemental leds once you get used to the cobs will help push you far past that 1gpw mark.