New Led Or Not ?

blacktourmaline said:

but for now The Epistars are great and they provide the widest wavelengths available

Click to expand...

You might want to check out Epistar's webpage.. they're absolute crap. 11,000mcd - 11 lumen. They don't make "3w diodes" as the panel makers claim.. not if they max out at 500mA.

http://www.epistar.com.tw/_english/01_product/01_overview.php

blacktourmaline said:

The only downfall i could see with DIY is the lack of available wavelength? Maybe someone can tell me some info about mixes different brands and wavelength to provide a good wide spectrum. As with using all White diodes I would imagine you will need at least some supplementary wavelength, but i bet you could use mostly white and they would perform good because of there wide spread. Also a White LED start out as blue but they add a phosphor coating.

Click to expand...

I already stated everything starts out as blue. And the stuff added is under the dome, not a "coating".

As for added supplementary wavelengths needed with the whites.. we need: 450nm, 470nm, 630nm, 660nm, and 735nm. The whites cover that quite well. Look at the Hortilux graph and compare. You'll see the whites have more than the HPS. Why supplement what you don't need?

As for lack of wavelength available for a DIY... why? Anything available to the factory is available to the public. Order it yourself.. google is your friend.

Spuzzum said:

Just figured it out... ALL diodes start out as blue.. ALL of them.

What does blue + red equal? Purple/violet.

http://en.wikipedia.org/wiki/Purple


That's a half-assed way to get violet... and waaaaay cheaper I might add. UV's are usually $10 and up.. some are $30.

Click to expand...

if thats true and they just change the color with a red lens, that explains why led´s suck at flowering and are good at veggin..

changing the color with a lens...

kinda like using an mh with a red lens, and claiming it does as good a job as an hps..

Spuzzum said:

You might want to check out Epistar's webpage.. they're absolute crap. 11,000mcd - 11 lumen. They don't make "3w diodes" as the panel makers claim.. not if they max out at 500mA.

http://www.epistar.com.tw/_english/01_product/01_overview.php

Click to expand...

So that full spectrum white led panel called "isunlight" says it uses all 3w epistars is a complete lie........lol
you should check out their prices.......its laughable....and epistar white diodes suck ass too......

Here's some of my Cree Whites.. just to compare...




CoolWhites are the R5's, and WarmWhites are the Q4's.. best cutter had on hand when I ordered.


CoolWhites:




WarmWhites:





Clear dome.

sso said:

if thats true and they just change the color with a red lens, that explains why led´s suck at flowering and are good at veggin..

changing the color with a lens...

kinda like using an mh with a red lens, and claiming it does as good a job as an hps..

Click to expand...

That's just panels made with ChinaCrap. There's some pretty decent grows out there with DIY Cree and Osram panels. Quality trumps price any day.


Just look at the guy at ledgrow.eu.. his setup 2 or 3 grows ago was better. He had top "and" side lighting, but now he's trying to sell them.. so doing just top lighting. But at only 60 or 90w, he's pulling some decent bud for the area.


Shit.. just look at "this" thread.. he does it like knna, and he's got some pretty nice results...

http://www.icmag.com/ic/showthread.php?t=158792

That's all that's needed.. reroute the damn traces to keep 3.xV and 2.xV separated. If need be, they can even use jumpers, or a double sided pcb.. ie: traces on both sides of the pcb, so there's no shorting by intersecting traces.

Pretty damn easy.. but cheaper if you don't.

Eraserhead said:

Click to expand...

OK.. looking at this pic again, I may have jumped the gun...

Now that I see Eraserhead's labeling of the diodes, then it's not "so" bad.. all reds on 1 string, and all blues/whites/violets on the other string. I say not so bad.. as that's still a difference in voltages, but all reds will be 2.x, and all blues/whites/violets will be 3.x.. with x equaling between 0.1 and 0.3v or so. I'm not sure if that's enough to cause them to die.. 0.1v, maybe not.. 0.2/0.3v.. at 500-700mA.. "maybe". 0.3V * 500mA = 150mW. (* 700mA = 210mW). Talking cheap ChinaCrap.. and if they haven't reflowed the diodes so the underside of the diode is soldered to a heatsink pad on the pcb.. then that makes a big difference as well.


I just kept seeing that "red" 410nm and thinking it's a red in the string with blues.

Each of those circuit board strips has 25 LEDs, and there are all mixed colors, I thought Eshine would have a better way of regulating what power goes where.

1 circuit board = 1 power supply.

The violets, blues and whites are ideally powered at 3.4 v @ 550mA, oranges, reds are ideally powered at 2.4v @ 550mA. So what happens, the reds and oranges are slightly over powered, and the blues and whites are slightly under powered, if there were an even amount of reds and blues on the same 25pcs strip, you'd get 2.9v @ 550mA. The reds and oranges max out at 3v and 1000mA, and the whites and blues max out at 3.8v and 1100mA.

So red and blue can be mixed on a string (just not optimally) but when you add 730nm or 840nm, we get a problem, and Eshine told me this is a normal issue with them. I have a remedy for that.

730nm and 840nm are ideally powered at 1.7v and 550mA, but can max out at 2.4v and 1000mA.

Spuzzum said:

OK.. looking at this pic again, I may have jumped the gun...

Now that I see Eraserhead's labeling of the diodes, then it's not "so" bad.. all reds on 1 string, and all blues/whites/violets on the other string. I say not so bad.. as that's still a difference in voltages, but all reds will be 2.x, and all blues/whites/violets will be 3.x.. with x equaling between 0.1 and 0.3v or so. I'm not sure if that's enough to cause them to die.. 0.1v, maybe not.. 0.2/0.3v.. at 500-700mA.. "maybe". 0.3V * 500mA = 150mW. (* 700mA = 210mW). Talking cheap ChinaCrap.. and if they haven't reflowed the diodes so the underside of the diode is soldered to a heatsink pad on the pcb.. then that makes a big difference as well.


I just kept seeing that "red" 410nm and thinking it's a red in the string with blues.

Click to expand...

Eraserhead said:

Each of those circuit board strips has 25 LEDs, and there are all mixed colors, I thought Eshine would have a better way of regulating what power goes where.

1 circuit board = 1 power supply.

The violets, blues and whites are ideally powered at 3.4 v @ 550mA, oranges, reds are ideally powered at 2.4v @ 550mA. So what happens, the reds and oranges are slightly over powered, and the blues and whites are slightly under powered, if there were an even amount of reds and blues on the same 25pcs strip, you'd get 2.9v @ 550mA. The reds and oranges max out at 3v and 1000mA, and the whites and blues max out at 3.8v and 1100mA.

So red and blue can be mixed on a string (just not optimally) but when you add 730nm or 840nm, we get a problem, and Eshine told me this is a normal issue with them. I have a remedy for that.

730nm and 840nm are ideally powered at 1.7v and 550mA, but can max out at 2.4v and 1000mA.

Click to expand...

E.shine may do it, bet every guide I've read on LEDs, and my own buddy who's guided me along the way.. who's an electrical wizard.. all say don't mix voltages. It may work with 1/4 watt 5mm LEDs, but not a good idea with high powered LEDs.

The way I've learned it... a 3.4v LED "needs" 3.4V just to light up. 2.4V will not do it. Reason being.. LEDs are "current driven".. not "voltage driven". It's the current that makes them go brighter or dimmer.. but you need that base voltage to even begin. So.. if a 3.4v LED is lit on a string with 2.4v LEDs.. then there must be at least 3.4v pumping through that string.

If you have a voltage meter, you can check really quick.. just measure a red and blue within the same string. See what they're running at.

I already have one on the way, along with a few other toys.

Spuzzum said:

If you have a voltage meter, you can check really quick.. just measure a red and blue within the same string. See what they're running at.

Click to expand...

I'm not too familiar with either Lumigrow or Cali Light Works. I've seen them online before and saw a couple videos, Lumigrow has like a 30 minute infomercial, but never seen any grows with either of them.

PSUAGRO. said:

Long story short and full of BS I got my shipping label emailed to me from Ben and no restocking fee.......Winning!!ha
Thanks Eraser....told him I'm investigating(bringing electrical engineer over/lol) whether his panels actually contain ANYTHING other than chinese diodes.....he said its slander/can take me to court..ha..this went on till this morning where he said that just don't say anything about this on the forums and ill get every penny back....oops

So back to the drawing board.........I've narrowed down to two companies Lumigrow(es165) and California Lightworks solarflare.. I'd like both of your opinions on these panels..Both are 100% built in America /both claim 5w top bin leds /massive heatsinks in both/ decent warranties....thoughts???

Strange both panels actual power outputs are 165w ......but Lumigrow panel claims 26diodes and cali panel claims 40 diodes....huh???

Click to expand...

OK, I stand corrected.. and will eat my words


Seems you CAN mix voltages on the same string, as long as there's enough voltage for all of them.. and there's either an adequate resistor at the end of the string, or you use a constant current driver. BUT.. the LEDs will all run at whatever the current is set to, so don't go over what the maximum current of the weakest diode.

http://www.greenpinelane.com/yaf/default.aspx?g=posts&t=424



What I had been reading.. and had stuck in my head.. was actually meant for LEDs wired in "series-parallel", or just plain parallel.




What happens here, is that all strings pull the same voltage, but you need 4x the current needed for "1" string, to supply the same amount of current to "all" strings. 5 strings would be 5x, etc...

Now.. if 1 single bulb blows, causing 1 single string to fail.. that 1/4 or 1/5 the current that's now "extra".. is then redistributed evenly amongst the remaining strings. That's where the diodes in the other strings will also blow if they can't handle the increase.

Makes sense.

http://www.kpsec.freeuk.com/components/led.htm#series



So.. I'm kinda p!ssed at my buddy now . Oh well... It works out with my controller.. I can dim each colour separately just by setting it's percentage with the press of a few buttons.



I'm honest.. if I make a mistake, I admit it.. that's how we learn.

Eshine power supplies runs at 60-90v, and the current is 550mA-700mA.

More efficient at 550.

I'm kinda curious how it actually works out if there are mixed voltages on the same line. There is 1.7, 2.4 and 3.4 voltage LEDs they use.

The way Eshine explains it, each color will draw only the voltage they need, and the rest of the available power goes to the remaining LEDs on that string.

I'm no electrician, so I can't say. All I know is, + and -. And don't touch electricity, because it hurts.

Eraserhead said:

Eshine power supplies runs at 60-90v, and the current is 550mA-700mA.

More efficient at 550.

I'm kinda curious how it actually works out if there are mixed voltages on the same line. There is 1.7, 2.4 and 3.4 voltage LEDs they use.

The way Eshine explains it, each color will draw only the voltage they need, and the rest of the available power goes to the remaining LEDs on that string.

I'm no electrician, so I can't say. All I know is, + and -. And don't touch electricity, because it hurts.

Click to expand...

Yep.. LEDs are current driven.. whatever extra voltage just flows straight through to the next LED, but still needs that initial base voltage. I've actually seen schematics where they use a red LED to drop the voltage by 1.7v or whatever the red uses, to bring a signal from 5v to 3-3.3v.

I even asked my buddy if I could stick a red at the end of a blue string, as it would've helped me use up the last few volts of a string.. making the drivers more efficient. He got a little ticked at me for continually asking what he already told me couldn't be done. Blah. Buddy's still a gold mine of information.. he's just older and grumpier than me...

Christmas lights are a perfect example... mixed colours/voltages on the same string.



But, back to why the diodes die...

They say the drivers are 550-700mA. They say they use Epistar 660nm's. Never mind that Epistar's 650nm (not 660nm) maxes out at 30mA, but a 624nm red maxes out at 500mA. They make blues that max out at 700mA, but if they list Bridgelux, my guess is they'd use Bridgelux blues over Epistar blues.

http://www.epistar.com.tw/_english/01_product/07_search.php

With Eshine and others similar from China, you have to forget about Epistar and Bridgelux, and not look to those websites for data on the Eshine, because we already proved they are not the LEDs they use. If you tell them you like Reebok LED lights, they will tell you they use Reebok LEDs. And not even think twice to see if we even made up the name or not.

You've seen some of my chats to the Chinese Spuzz, I think I will contact them and see if they can put some "Reebok LEDs" in a fixture for me. This could be fun

Spuzzum said:

Yep.. LEDs are current driven.. whatever extra voltage just flows straight through to the next LED, but still needs that initial base voltage. I've actually seen schematics where they use a red LED to drop the voltage by 1.7v or whatever the red uses, to bring a signal from 5v to 3-3.3v.

I even asked my buddy if I could stick a red at the end of a blue string, as it would've helped me use up the last few volts of a string.. making the drivers more efficient. He got a little ticked at me for continually asking what he already told me couldn't be done. Blah. Buddy's still a gold mine of information.. he's just older and grumpier than me...

Christmas lights are a perfect example... mixed colours/voltages on the same string.



But, back to why the diodes die...

They say the drivers are 550-700mA. They say they use Epistar 660nm's. Never mind that Epistar's 650nm (not 660nm) maxes out at 30mA, but a 624nm red maxes out at 500mA. They make blues that max out at 700mA, but if they list Bridgelux, my guess is they'd use Bridgelux blues over Epistar blues.

http://www.epistar.com.tw/_english/01_product/07_search.php

Click to expand...

lol!


Counterfeit Epsitar's? That's pretty bad.. counterfeiting a cheap-assed diode to begin with, that still lists in mcd

I figured they just used cheap Epistars, and pushed them to the absolute limits.

I still pull Plenty of weight off the epistars so im happy ......Where all the BIG grows with these DIY leds, I wanna see some big plants with phat nuggets!!! all i see is PC boxes and autoflowers.....guess it cost way to much to make one compared to a 1000w....

Has anyone else noticed that Hydroponics Hut removed all references and logos regarding Cree from the website?

I think with the combination of me emailing him and putting a little fire under his ass about fake Cree's, and Psuagro getting on him for being shady, he realized it is best to just tell the truth.

And btw,

Advanced LED removed all references to Bridgelux and Epistar from the Diamond Series description as well. Not sure if it had to do with my emails to him asking him to explain how he has spectrums from specific companies that they do not even make.

People are getting wise to the tomfooleries from these LED companies.