Analog's LED

[PetFlora]

FYI, the kelvin range does not tell you the output of the light source. It doesn't help much to give you an idea of the quality of the light.

You can't compare LEDs to CFLs because LEDs have a very different type of light output. White Leds have a much much better quality of light than CFLs, that's why you are seeing better results under the Leds.

Floro's emit "peaks" of wavelengths at a few wavelengths and they are really not ideal peaks for growing. Yes, they do work somewhat since plants can adapt to crappy light sources.

Leds on the other hand emit a "wide-spectrum" type of light. If you take a look at a white led spectral analysis over a cfl spectral analysis you will see what I'm talking about. Leds cover a broad range of wavelengths as a gradual curve from blue to red.

Yes, white leds will always grow plants better than crappy floroescents do.

I know you know this, so chalking your reply up to being high, and not clarifying what you mean-lol.

Not ALL LEDs emit wide spectrum- only whites. And not all cfls are crappy. HO T5 aquarium bulbs rock it just fine + plus they cover considerably bigger footprint for < half the price of today's cost of quality neutral/warm white LEDs.

That said, you now I am a fan of where LEDs are finally at.

Analogue, how big a footprint are the whites throwing, and could they have been a bit farther apart?

 

[analogue]

footprint isn't much larger than the heatsink. but i also need some area with less light, for fresh cuttings.

if i had ordered all my components locally i would have been better off.

i found out the solderless wire interconnects are not compatible between ModularLED and RapidLEd. drat!

hindsight is 20/20

 

[tenthirty]

Analogue,

How much money do you have involved in your little led experiment?

 

[analogue]

To be honest I'm not keeping a running total on costs.

My better half is enthusiastic about our lower electricity bill.

I like to design and plan lights to see what the bare costs are. Primary expenses are heatsink, diodes and drivers. Minimal expenses for little things like lenses, connectors, wires, hanging hardware.

Others, such as Rives, are correct in saying that the expense is about the same. I think having ownership of how it works is pretty handy.

 

[medmaker420]

analogue - yes I think people are looking for this wonder drug sort of light that will consume 55 watts and grow a pound. That is not what leds can do and if it could it sure as hell wouldn't be cheap to say the least lol.

 

[Wizzi]

OT sorry guys, I need some advice on buying an LED panel, what do you think of this:http://www.ledhydroponics.co.uk/OnlineShop/tabid/88/ProdID/24/Quantum_Photon_PRO_126_.aspx

ps:scusate per l'inglese

 

[analogue]

That company looks like a reseller of led light clones fr.om China. In the last week I've lost count of new websites selling the same exact things with different names.

Will it work? Probably. For how long? Your guess is as good as mine.

 

[analogue]

PetFlora: do you remember if SX mentioned or discussed a ratio of Cree Neutral White to 630nm Red? example: 24 leds, 18 red, 6 white, or the other way around .

Was looking for a reference but can't find one...

 

[tenthirty]

PetFlora: do you remember if SX mentioned or discussed a ratio of Cree Neutral White to 630nm Red? example: 24 leds, 18 red, 6 white, or the other way around .

Was looking for a reference but can't find one...

Please correct me if I'm wrong, but I'm thinking 10 XM-L to 8 XP-E leds.
This is based on power supply constraints and IMHO the difference in output from the 2 different diodes.

 

[analogue]

Interesting. I would like to see solderless versions of the Neutral White XM-L's. They may be pre-tinned for easier soldering but I don't have the skills for that yet (trying). Using those XM-L chips with this driver would be awesome.

 

[tenthirty]

Interesting. I would like to see solderless versions of the Neutral White XM-L's. They may be pre-tinned for easier soldering but I don't have the skills for that yet (trying). Using those XM-L chips with this driver would be awesome.

Yupers, That's the driver.

I'm thinking a 36"x1.5" wide heat sink, 18 diodes, remote power supply s, passive cooling. About 100w per panel. 3'x1' coverage. Remember Mr.X was using fewer watts with older diodes and look at what he was able to produce.

I'm thinking about 300w over a 3x3 square is a good starting point with available hardware and not having to drill holes or do anything custom. For me soldering is cake and I do have a complete machine shop, but for prototyping time is everything.

If you need help with soldering, PM me and we'll talk.

 

[analogue]

Sounds like a good plan - I hope you journal it.

---

Both of these are Harle-Tsu, from seed.

 

[PetFlora]

PetFlora: do you remember if SX mentioned or discussed a ratio of Cree Neutral White to 630nm Red? example: 24 leds, 18 red, 6 white, or the other way around .

Was looking for a reference but can't find one...

Scroll down ~ 1/2 way. My interpretation: NWs will do the whole thing, but I would add ~ 10% 660 reds on a separate circuit/timer. Ditto IR

https://www.icmag.com/ic/showthread.php?t=229278&page=5

 

[PetFlora]

Interesting. I would like to see solderless versions of the Neutral White XM-L's. They may be pre-tinned for easier soldering but I don't have the skills for that yet (trying). Using those XM-L chips with this driver would be awesome.

www.rapidled has Cree NWs and solderless kits- basically plug-n-play. PLUS, they offer pre-tapped/drilled heat sinks. SWEET

 

[PetFlora]

Interesting. I would like to see solderless versions of the Neutral White XM-L's. They may be pre-tinned for easier soldering but I don't have the skills for that yet (trying). Using those XM-L chips with this driver would be awesome.

They offer dimmable and non dimmable drivers

 

[PetFlora]

Yupers, That's the driver.

I'm thinking a 36"x1.5" wide heat sink, 18 diodes, remote power supply s, passive cooling. About 100w per panel. 3'x1' coverage. Remember Mr.X was using fewer watts with older diodes and look at what he was able to produce.

I'm thinking about 300w over a 3x3 square is a good starting point with available hardware and not having to drill holes or do anything custom. For me soldering is cake and I do have a complete machine shop, but for prototyping time is everything.

If you need help with soldering, PM me and we'll talk.

I think passive heat dissipation could be a problem with 18 diodes on 1.5" wide heat sink. Rapid can help with that

 

[analogue]

Scroll down ~ 1/2 way. My interpretation: NWs will do the whole thing, but I would add ~ 10% 660 reds on a separate circuit/timer. Ditto IR

https://www.icmag.com/ic/showthread.php?t=229278&page=5

The specific information about how many Neutral Whites SX would use with Reds is not there on that page.

He talks about a lot of things but not ratios of Neutral White to Red.

www.rapidled has Cree NWs and solderless kits- basically plug-n-play. PLUS, they offer pre-tapped/drilled heat sinks. SWEET

We know, thanks.

They offer dimmable and non dimmable drivers

We have already made more than one light with drivers from them.

I think passive heat dissipation could be a problem with 18 diodes on 1.5" wide heat sink. Rapid can help with that

It won't be a problem. I don't see how it could be an issue. Rapid spaces their drill/tap sites 2 inches apart. One of the reasons we use Rapid for Heatsinks is their slotted fins are designed to allow for the optional use of fans. Cool!


----

There isn't a reason to have your 630 or 660nm reds on a separate timer. By default you should have them on a separate circuit since their ceiling is at much lower mA's.

As for as IR (740nm), there's already post 700 nm in the Cree Neutral whites. That's one of the reaons We like them so much.

Running additional IR (hey, it could be Halogen) on a separate timer after sundown / lights-off, is not required. If you are having problems with flower initiation - maybe, but not before looking at your photoperiod first.

 

[tenthirty]

Remember the XM-L diodes will be driven at about 6w and the XP-E's will be at about a quarter of that. 1700ma vs 350ma more or less.

When I get some time, I'll start a thread and I'll site previous art and data as well as my logic on what, why and how.

Thanks all!!

 

[vukman]

http://donklipstein.com/photosyn.html
http://donklipstein.com/ledfaq.html#gr

Some rather interesting reading material on the subject being discussed.. I posted this in the botany forum but just in case 'y'all' haven't been there...

 

[tenthirty]

Only being able to drive the reds at 1w, that gives me about 60w of white and 9w of red.
Unfortunately, reading the datasheets, it appears to me that there really is no way to correlate reds to whites, short of a quantum meter.

69w of light per bar hits me as being a little shy of what I'm looking for to cover a 1x3' spot for flower, but on the good side, there is that much less heat to dissipate over the heat sink. So maybe each bar will be about the same power as the EVO led systems.

If you haven't seen Mr.X
https://www.icmag.com/ic/showthread.php?t=215147

From an engineering point of view, I'd rather have to deal with a lack of light apposed to a lack of cooling.

As the posts above discusses, They only talk about discrete (more or less) colors or frequencies. IMHO I think that there is some kind of synergy going on with a fullish spectrum. I have yet to find any data on resonance or hetrodyning and what effect this would have if any. By the way, this is not my idea. Knna brought this up in some post somewhere, that I had stumbled across long ago.

The effects of mixing and matching frequencies in the audio and radio bands is well know and documented in various life forms and electronics. I just have to wonder if the same would hold true for light utilization, be it the wave or the particle, or both that is being harnessed by the plant. After all, they have been doing it for a couple of billion years. It stands to reason that a plant uses as many of the properties of light that it can.