Chinese -Leds

[sethimus]

That is some high electrical costs you have there. But even with your calcs I would pick the epistars, 3.65 years, I will be upgrading to better LEDs by then

see, that's you. you probably like to litter the planet. i don't. and i would never assume the difference in these lights is only 200w, that's the model calculation to get nice even numbers. we've seen diy grows here in the 1.5g/w range already, in reality the break even numbers are way lower

and why would you switch to "better leds" by then? just fork over that money in the first run and get a better led right now.

 

[Yodaweed]

Where's that dead horse again....I think he might need a lil more beatin.

 

[nomofatum]

see, that's you. you probably like to litter the planet. i don't. and i would never assume the difference in these lights is only 200w, that's the model calculation to get nice even numbers. we've seen diy grows here in the 1.5g/w range already, in reality the break even numbers are way lower

and why would you switch to "better leds" by then? just fork over that money in the first run and get a better led right now.

The best LED available now (without insane cooling) will get 130-150 lumens per watt. A led from 3 years from now will put out well over 200 lumens per watt. If you care about the planet you will upgrade before burning all that extra fuel to power your lights.

My plan:
105lm/watt 2.5 years.
200lm/watt 2.5 years

Your plan:
130lm/watt 5 years

My average is higher than yours, and I spent less on LEDs by buying the 2nd best option twice instead of the best option once.

 

[sethimus]

The best LED available now (without insane cooling) will get 130-150 lumens per watt. A led from 3 years from now will put out well over 200 lumens per watt. If you care about the planet you will upgrade before burning all that extra fuel to power your lights.

but you want to buy the cheap tech from 3 years ago and older. in 3 years you would get the tech from today cheap, not the new efficient stuff.

and the cheap lights you so love never ever get 130-150lm/w, that's cob territory

 

[sethimus]

The best LED available now (without insane cooling) will get 130-150 lumens per watt. A led from 3 years from now will put out well over 200 lumens per watt. If you care about the planet you will upgrade before burning all that extra fuel to power your lights.

My plan:
105lm/watt 2.5 years.
200lm/watt 2.5 years

Your plan:
130lm/watt 5 years

My average is higher than yours, and I spent less on LEDs by buying the 2nd best option twice instead of the best option once.

no, i will go the diy route (no usable room right now), and just switch cobs after a few years with newer ones and reuse all the other parts, the beauty of diy. you will throw out your shit lights every 2.5 years. just like the hid crowd their bulbs now, only with a longer cycle.

and i highly doubt you will get a 200lm/w light in 2.5 years that cost as much as a shit light today.

 

[nomofatum]

no, i will go the diy route (no usable room right now), and just switch cobs after a few years with newer ones and reuse all the other parts, the beauty of diy. you will throw out your shit lights every 2.5 years. just like the hid crowd their bulbs now, only with a longer cycle.

and i highly doubt you will get a 200lm/w light in 2.5 years that cost as much as a shit light today.

Check the news. Cree, Epistar, and a few others already have >200lm/w chips coming out this (2015) year. 200lm/w won't be a problem in 2.5 years, there will be many options.

 

[sethimus]

Check the news. Cree, Epistar, and a few others already have >200lm/w chips coming out this (2015) year. 200lm/w won't be a problem in 2.5 years, there will be many options.

the cxa 3070 is now avaiable for nearly 2 years and i have not yet seen one commercial light using it. instead the shit light makers still insist on littering the market with old tech, why would they change the formula when ppl like you are happy to buy their stuff?

 

[littlejacob]

Hello weeders!!!

I'm French(I know,nobody's perfect...)and I like to know what kind of spectrum do you use?
I join a picture of one of my 54w(18*3)led spot that I buy few weeks ago cause a guy from a French forum,where I'm a user,made 2.4g/w with 4 in vegging and 6 in flowering.Their are 20 real watt...but the guy is 54 years old and with hps he was always at 1.2,1.3 g/w under 600w and scrog
If anyone want the link to the grow log no problem but it's in French.
Good grow everybody!

 

[nomofatum]

Hello weeders!!!

I'm French(I know,nobody's perfect...)and I like to know what kind of spectrum do you use?
I join a picture of one of my 54w(18*3)led spot that I buy few weeks ago cause a guy from a French forum,where I'm a user,made 2.4g/w with 4 in vegging and 6 in flowering.Their are 20 real watt...but the guy is 54 years old and with hps he was always at 1.2,1.3 g/w under 600w and scrog
If anyone want the link to the grow log no problem but it's in French.
Good grow everybody!

Google can translate for me, link please.

 

[Yodaweed]

Hello weeders!!!

I'm French(I know,nobody's perfect...)and I like to know what kind of spectrum do you use?
I join a picture of one of my 54w(18*3)led spot that I buy few weeks ago cause a guy from a French forum,where I'm a user,made 2.4g/w with 4 in vegging and 6 in flowering.Their are 20 real watt...but the guy is 54 years old and with hps he was always at 1.2,1.3 g/w under 600w and scrog
If anyone want the link to the grow log no problem but it's in French.
Good grow everybody!

Very impressive @littlejacob got any pictures of the buds? What strain was it? I prefer a predominately white and red spectrum myself but sounds like what you got going is pretty good.

 

[captainmorgan]

2.4 g/w is a huge number and hard to believe out any conventional set up,I've seen claims with similar numbers but they were with a very specialized hydro set up. I've pulled 1.3 to 1.4 in 5 gallons of dirt and the plant needed every branch supported because of that weight. I would like to see a link.

 

[littlejacob]

Ok guys
The web site is : Cannaweed.com and search for:jdc Yannos, s.t.g, led test on line, scrog. It's bionicled a french led seller who gave him the lamp to test it online, they were sure of their spectrum...and they were right.The same guy gonna test their cob led soon and this guy is on the site since a long time so he's not working for them.
See you soon

 

[littlejacob]

Hello again weederz!
If someone want to know there is 18 epistar led in my spot run at 350mhA,2x440nm,2x460nm,1x610nm,1x730nm in the middle and on side 8x660nm +4x630nm and I believe that one lamp can cover 1sqft for the best result. If you want more detail go to bionicled.com but I think it's only in french
See you M.J lovers

 

[kmog33]

Chinese leds.

Sent from my LG-LS980 using Rollitup mobile app

 

[Positivity]

^^

330w ti smartlamp AND a 400 hps in there?

 

[SupraSPL]

IMHO it's a choice of DIY Cree vs DIY Epistar. The Cree is the better product, but it costs about 8x to 10x the cost of Epistar. Eventually that cost difference will pay off, but the question is will you still be using it at that point. In my case, it would take over 5 years for the Crees to pay off, and I most certainly will not be using the same system in 5 years, especially with how much better they are getting so quickly.

Thanks for the link (warm white 30 watters for $1.84 ea). I will definitely grab a batch for testing. I think the main disagreement we have is that you believe the generics can provide 110lm/W at operating temp and current and I do not. But if you turn out to be correct, that is great news for all of us so I hope you are correct. At 110lm/W, I still would not use them for growing on a large scale, but probably very good for indoor lighting. A I said, I run my 3000K CXAs at 164lm/W did you look over the PDF and do you still disagree about the Cree's performance?

Even if the generics are only 18% efficient, that is $0.50/PAR W. Top bin Crees at 1.4A would cost about $1.77/PAR W and at 700mA they cost $3.25.PAR W. Vero29 at 2.4A cost $0.90/PAR W. I actually agree with your argument about how long it would take to make up that cost difference in terms of LED dissipation electricity. But the other associated costs of low efficiency are much much higher. I think you will learn that with experience. Primarily we are talking about heat in your grow space but the extra cost of heatsinks and drivers is substantial and must be factored in.

 

[SupraSPL]

I put together a 20W generic that I had on hand and put it on a $1 reclaimed heatsink. The ad says 100lm/W at max power. I though it might be good for indoor lighting because it is relatively diffuse. I have noticed that lighting indoors with Cree XTEs is too sharp on the eyes, it needs to be out of the line of sight, diffused or bouncing off of something. Anyway this 20 watt warm white generic COB is built on a piece of aluminum that is very very unflat, just like 20mm LED stars. We have learned that sanding the heatsink surface provides no benefit too cooling as long as you use a good thermal paste. But this is different, I had to sand it at 120-220grit for 10 minutes before the entire surface of the aluminum would even touch down. That amount labor already cost more than simply buying a perfectly flat COB like a $9 Vero 13, which really can do 110lm/W at 20W. Although if I had a belt sander set up it would be a very quick fix.

Anyway, if we ignored the non-flatness and used a ton of thermal paste with no air bubbles, I am sure it would cool well enough. Right away I noticed that there were chunks missing from the COB encapsulant and the chip under that spot does not light up. After the first use I noticed another of the chips inside stopped lighting up as well. At this point I know I made an $8.30 mistake LOL.




So this is a $4.5 Vero 10 2700K vs $8.30 generic 3500K.

 

[Positivity]

Oh sh@t it broke...

Add research time for weeding out the duds..

 

[harris hawk]

The best LED available now (without insane cooling) will get 130-150 lumens per watt. A led from 3 years from now will put out well over 200 lumens per watt. If you care about the planet you will upgrade before burning all that extra fuel to power your lights.

My plan:
105lm/watt 2.5 years.
200lm/watt 2.5 years

Your plan:
130lm/watt 5 years

My average is higher than yours, and I spent less on LEDs by buying the 2nd best option twice instead of the best option once.

What type of LED's do you recommend?

 

[SupraSPL]

I will be the first one to say that this test is far from perfect because of the varying color temps and the varying output patterns, but it can be used as a very general guide. I was able to control the other variables very well. The measurements were done pulsed at 23C, 5.25" away from the light sensor. LED dissipation was measured precisely and worked into the results.

Generic 3K bars @ 160mA: 1490lux/W (1.85W diss @~40% efficient = $4.40/PAR W)
Generic 3K bars @ 370mA: 1188lux/W (4.68W diss @~32% efficient = $2.17PAR W)
Generic 3.5K COB @ 300mA: 1250lux/W
Cree 3.2K XTE R3 @160mA: 1736lux/W
Cree 3.2K XTE R3 @370mA: 1497lux/W (130lm/W expected from PDF)
Cree 3.2K XTE R3 @1470mA: 834lux/W
Vero10 2.7K @ 150mA: 1871lux/W (135lm/W) (3.88W diss @ 42.9% = $2.80/PAR W)
Vero10 2.7K @ 370mA: 1610lux/W (115lm/W) (9.81W diss @36.5% = $1.29/PAR W)
Vero10 2.7K @ 1490mA: 962lux/W (69lm/W) (47.38W diss @ 21.8% efficient = $0.45/PAR W)
Vero29 3K @ 100mA: 2381lux/W
Vero29 3K @ 150mA: 2365lux/W
CXA3070 3K AB @100mA: 2522lux/W
CXA3070 3K AB @150mA: 2514lux/W
CXA3070 3K AB @370mA: 2417lux/W (188lm/W) (12.37W diss @ 57.8% = $5.60/PAR W)
13W 3K CFL with partial reflector: 441lux/W (warmed up)

The $1.50 generic bars performed surprisingly well compared to the top bin XTEs, although the XTEs are not exactly cutting edge for Cree these days. Despite the surprising performance of the generic bars, the Vero10 was able to beat them in cost by a large margin AND maintained a higher efficiency.


I put the large COB results in there just for the heck of it, but of course it is not fair because they are being under driven to the extreme. Personally, I don't mind paying $5.60/PAR W, I have paid much higher in the past for a lot lower efficiency.

Out of curiosity, at the end of the test I put 1.5A into the generic COB and it burnt out within 10 seconds, despite being perfectly cool. I guess there is only one way to find out how much is too much. The Vero10 does not seem to mind the 1.5A current other than losing half its output to current droop. I let it sit there until I could smell the outgassing and it was so hot I could not touch the heatsink for more than a split second (55C+). It maintained its output surprisingly well despite very insufficient cooling (-20% to temp droop). Of course if you were to use the Vero in a growing lamp using sufficient cooling it would run just fine and stay cool.