Another LED thread. Coming from HPS, looking at ALL options. Whats too much?

[doz]

**Wall of text, cliffs at the bottom**

A little background. I grow for enjoyment more than anything. It interests me and I have a good time doing it. My space is limited however which is the reason I am looking into LED from HPS. I know my area is NOT ideal but I am looking to get the most out of it.

I run a room that is 2' wide X 4.5' long X 7' tall. I used to have a 4x4 area, but had to downsize when I moved. I am still running the 1000w from then. This is NOT ideal. I was actually told by the local Hydroponics store owner that I would not even be able to yield 1/2 pound in my area and Id be lucky to finish a grow with that much light in a small area. He then politely told me I had no idea what I was doing and that growing is not for everyone! I did not want to lead on too much info and I did not have the heart to tell him that I yielded 21 ounces this past week in 9 sq ft (no CO2 used, need to invest in this). While my area is not ideal, I am pretty happy with the results thus far. The store owner told me that 1000w in my area is WAY too much light. Is there such thing as too much light? Has anyone walked outside lately? Id really like some opinions on this and how people feel about too much light. Minimum light seems to be 50w per sq/ft for HPS which would mean that I need roughly 500w MINIMUM. The owner said I would only need 400w for my area.

Getting past all of that, I am looking for LED options. I really want to run 2 LED lights to distribute the light a bit better for my area. If I dont go LED, I plan on going with 2x400w or 2x600w HPS. Ive read TONS of threads on multiple forums. Each forum seems to have their "favorite". I have read the what is to be believe as poor business practices by PlatinumLED. Seems they are not liked around these parts. I have read about MarsHydro lights and how they are pretty decent price to performance, but again its more for the "budget grower" (please correct me if I am wrong). Then there are the countless knock offs.

Where I am at is basically looking at AdvancedLED, PlatinumLED, and Lush Lighting. I know a lot of people dislike Mark from PLED for his practices. Can anyone comment about the actual light itself? He claims to use 3w Bridgelux LED just like the ALED. They (until recently) both use the same housing. To me it seems they are basically the same. PLED offers 5 year warranty vs. ALED has a 3 year warranty. ALED is quite a bit more expensive. Can anyone comment any further (please refrain from "shit talk", I would like honest opinions about the lights they have used or facts about the lights that they know).

And Lush. How do people feel about Lush lighting? Their offerings are about inline on pricing per watt with ALED (more expensive than PLED). Ive seen good results but they dont seem to be as popular.

Any info about other LED lights is welcome as well. Again, I am trying to do as much research as possible. I do not have a problem cooling my grow area. It sits about 75-78 degrees with the 1000w in there. I am just looking to move to something more per say. I do believe that LED can produce more resinous buds which means higher quality. I do want quality, but I enjoy the yields I am getting with HPS. I will accept a little lower yield for a higher quality though without complaint! Again, all information would be appreciated especially if anyone has FACTS about the difference between PLED and ALED. Ive read about Mark and his "attitude" but I am really after the QUALITY of his lights vs. ALED.

Cliffs:
Currently running 1000w in 9.5sq ft w/ avg yield around 20 ounce no CO2
Looking to upgrade to LED
Want FACTS about ALED vs PLED and info about Lush Lighting and other QUALITY LED fixtures
Would like info about what people feel is TOO MUCH LIGHT. Is there such thing as too much?

Pictures of the recent harvest. 21 ounces of Barney's Farm Critical Kush!

 

[bicit]

Well, I would agree that 1000w is a bit too much. Point of diminishing returns and all that(IE it may be possible to get the same results with a 600w hps). Are you willing to delve into the world of DIY lamps? You can get much better results and cut your power consumption by a considerable amount for less(or the same) money as commercial lamps.

Other wise I would look into companies like optic grow lights, Apachee tech, Area 51, or onyx LED lamps. They're the best commercial options available right now. I would stay far away from any of the lights you mentioned, especially the mars and it's various knockoffs.

 

[doz]

I have the means, although I am not great with electrical. I would not be completely opposed to it. I can make a metal fixture myself to mount everything with aluminum. I just really do not know enough about the technology and I know that diving in head first usually results in a bit of mishap and usually costly (although rewarding down the road).

And I do not doubt that 1000w is hitting the point of diminishing returns as I should have a 4x4 footprint rather than 2x4.5 but as I said, its what I had from a previous room. I do not believe its "too much" to the point where it will harm anything though.

My last three harvests have been 18, 19.75, and 21 ounces (first two were RP OG #18, last one CK).

 

[mc130p]

I have the means, although I am not great with electrical. I would not be completely opposed to it. I can make a metal fixture myself to mount everything with aluminum. I just really do not know enough about the technology and I know that diving in head first usually results in a bit of mishap and usually costly (although rewarding down the road).

And I do not doubt that 1000w is hitting the point of diminishing returns as I should have a 4x4 footprint rather than 2x4.5 but as I said, its what I had from a previous room. I do not believe its "too much" to the point where it will harm anything though.

My last three harvests have been 18, 19.75, and 21 ounces (first two were RP OG #18, last one CK).

It seems to me like these grow store guys don't know shit. I've always been too paranoid to go to one, but the stories I hear are horrendous. But, you can have too much light. It can lead to bleaching of bud tips, and curling or even dying leaves.

If you did a DIY, you could replace the HPS altogether with a setup that gives more light with less heat and power $$. If I had your space, I'd shoot for about 600W of CREE CXB total max.

There are a number of threads in here that have step by step builds, often with links to supplies. They might be for an older model array, but if you understand how those were done, you can easily use the new chips often with just straight substitution. The circuits are very simple DC series circuits and are pretty quick to learn. A space your size is really well-suited for a build that you customize yourself.

 

[apoulin]

It seems to me like these grow store guys don't know shit. I've always been too paranoid to go to one, but the stories I hear are horrendous. But, you can have too much light. It can lead to bleaching of bud tips, and curling or even dying leaves.

If you did a DIY, you could replace the HPS altogether with a setup that gives more light with less heat and power $$. If I had your space, I'd shoot for about 600W of CREE CXB total max.

There are a number of threads in here that have step by step builds, often with links to supplies. They might be for an older model array, but if you understand how those were done, you can easily use the new chips often with just straight substitution. The circuits are very simple DC series circuits and are pretty quick to learn. A space your size is really well-suited for a build that you customize yourself.

Well put

 

[nogod_]

Platinum probably doesnt use bridgelux, they almost certainly use epistar. Mars also uses epistar. Lush is a rebranded mars. In short if you want a cheap premade unit the mars [~25% efficient] will grow but youd probably be better off just using a 600w hps [~40% efficient].

As far as amount of light....most folks are using somewhere between 25-40 watts sq/ft. I know church uses something like 18.

Building a diy lamp is fun. Ive built a couple different modules with absolutely zero experience whatsoever. You get to customize your footprint, your spectrum, your efficiency, and if something breaks you can fix it yourself in minutes if you keep spare parts on hand.

 

[churchhaze]

Platinum probably doesnt use bridgelux, they almost certainly use epistar. Mars also uses epistar. Lush is a rebranded mars. In short if you want a cheap premade unit the mars [~25% efficient] will grow but youd probably be better off just using a 600w hps [~40% efficient].

As far as amount of light....most folks are using somewhere between 25-40 watts sq/ft. I know church uses something like 18.

Building a diy lamp is fun. Ive built a couple different modules with absolutely zero experience whatsoever. You get to customize your footprint, your spectrum, your efficiency, and if something breaks you can fix it yourself in minutes if you keep spare parts on hand.

The newest cxb build is ~193W dissipated over 8sqft, so about 24W/sqft. At 49.1% efficiency, that's about 11.8 PAR W/sqft.

The vero 18 build is 240W dissipated fo 8sqft in an identical tent, so 30W/sqft. At about 40% efficiency, that's about 12 PAR W/sqft.

I think 12 PAR W/sqft "adequate" for overhead lighting for 3 foot plants. 14 PAR W/sqft is very good. 12 is nothing to sneeze at. It's probably perfect for 2.5 foot tall plants.

 

[alesh]

I'd aim at 600W of quality DIY LEDs (CXBs) or stay with the HPS as you seem to have great results with it. Don't expect better results with LEDs as there's not much room for bigger yield in your space.

 

[Doer]

Right, if you put down 600w, you need 1500w of LED power going in at 38% efficiency....if you can get that.

And according to the expense up front, you can choose how many COBs, that is to drive them softly, and get a reasonable trade off at a cost point you can live with.

Then add the drivers to supply the juice. That has to be calculated from how much amps you intend to drive at what volts. Less is better, for the most part, but it means more emitters and drivers to do the same job.

All trade offs.

We call this in enterprise computing, calculating the Total Cost of Ownership.

Some folks are fine with the TCO of a mega-yacht. We were fine with the TCO of Saturn V, at the time. But, it could only lift 4 inches per gallon of fuel.

 

[alesh]

Right, if you put down 600w, you need 1500w of LED power going in at 38% efficiency....if you can get that.

And according to the expense up front, you can choose how many COBs, that is to drive them softly, and get a reasonable trade off at a cost point you can live with.

Then add the drivers to supply the juice. That has to be calculated from how much amps you intend to drive at what volts. Less is better, for the most part, but it means more emitters and drivers to do the same job.

All trade offs.

We call this in enterprise computing, calculating the Total Cost of Ownership.

Some folks are fine with the TCO of a mega-yacht. We were fine with the TCO of Saturn V, at the time. But, it could only lift 4 inches per gallon of fuel.

The cost of COBs is stacking up when run soft, the cost of drivers is usually not.

 

[Doer]

The cost of COBs is stacking up when run soft, the cost of drivers is usually not.

The cost of drivers is independent. Again these are just choices.

 

[doz]

$40 per LED which according to digikey.com. Is that right for cxa3070?? You figure (according specs) they will run at about 100w each at 40% (rough numbers) of actual lighting meaning you would need 6 (figure 3 per fixture, 2 figures to disperse the light across the room properly). So looking at $240 for the COBs correct?

Then I would need to either buy or make a fixture, and add heatsinks and fans to keep her cool. I have not had a ton of time to read but from what I have read, the COBS can be driven just from the 110v power outlet? So youd be looking at probably another $200 for parts (if no drivers are needed). About $400-$500 for a 600w ACTUAL LED fixture which is really not bad at all.

My question then would be what to do about spectrums? With the COBs, they seem to only sell 2700k, 3500k, 4000k, 5000k (I think??). Do you add a few LED singles just for difference spectrums?

Ill definitely need to do some more reading this evening/week when I have time. Seems pretty interested although a little daunting at first as is anything. Half tempted to spend $600 on a dual 400w HPS setup to run for the next year while I experiment with LED tech.

 

[churchhaze]

I would not listen to the obvious LED troll. (not doz)

You should design around a particular driver you choose, not the other way around. It's better to start with an HLG-185H-C1050, for example, then scale around multiples of cobs that works best with that driver's voltage range.

The closer you can get driver load to 100% with these drivers, the better. Not only do you get better value for your drivers, you also get higher efficiency.

If you limit your choices to HLG-185H-C and HLG-120H-C models, it really narrows down the possible choices and makes design more about picking drive current and what set of multiplies is best for your design. Those 2 drivers have 94% efficiency at full load, and active power factor correction.

Like @alesh was saying, the cost of drivers obviously doesn't stack with up like the initial cost of buying extra cobs and running them softer. This different in price scaling should definitely be taken into consideration. The price of heat sinks doesn't stack when you under drive with more cobs either, in fact it allows you to reduce the total heat sink size.

 

[Rahz]

I think the yield you mentioned could be achieved and surpassed with a DIY build of 500W Total watts, 400-450W of actual output.

 

[bicit]

$40 per LED which according to digikey.com. Is that right for cxa3070?? You figure (according specs) they will run at about 100w each at 40% (rough numbers) of actual lighting meaning you would need 6 (figure 3 per fixture, 2 figures to disperse the light across the room properly). So looking at $240 for the COBs correct?

I'm guessing you're talking about the 3000k, Z2 bin. Which is listed at ~73w if driven at 1.9amps and it would be roughly 33.5% efficient, producing 24.5 par/w or $1.63/parw for the bare emitter. The vero 29 is much cheaper and more effecient at this power level (~38% efficient at 79W.). The new vero 2.0 and CREE CXB series are going to bump those numbers higher still.

Then I would need to either buy or make a fixture, and add heatsinks and fans to keep her cool. I have not had a ton of time to read but from what I have read, the COBS can be driven just from the 110v power outlet? So youd be looking at probably another $200 for parts (if no drivers are needed). About $400-$500 for a 600w ACTUAL LED fixture which is really not bad at all.

No cobs cannot be driving off AC. They still need a constant current driver to provide DC. I would figure closer to $900 for 600w, but it depends on how you build it.

http://rollitup.org/t/diy-leds-how-to-power-them.801554/
http://rollitup.org/t/heatsinks-for-diy-led-lamps.851907/
https://www.rollitup.org/t/diy-led-cree-cxa3070.789575/

My question then would be what to do about spectrums? With the COBs, they seem to only sell 2700k, 3500k, 4000k, 5000k (I think??). Do you add a few LED singles just for difference spectrums?

General rule of thumb around here, 3000k for flower, 5000k for veg. Everyone has slightly different preferences though. (really getting photons to the plant is the most important thing)

Ill definitely need to do some more reading this evening/week when I have time. Seems pretty interested although a little daunting at first as is anything. Half tempted to spend $600 on a dual 400w HPS setup to run for the next year while I experiment with LED tech.

That seems like a lot of money to spend on a 400w HPS. I'd look into the 315w Philips ceramic metal halides. Seems like a better choice for smaller spaces.

 

[medicinehuman]

$40 per LED which according to digikey.com. Is that right for cxa3070?? You figure (according specs) they will run at about 100w each at 40% (rough numbers) of actual lighting meaning you would need 6 (figure 3 per fixture, 2 figures to disperse the light across the room properly). So looking at $240 for the COBs correct?

Then I would need to either buy or make a fixture, and add heatsinks and fans to keep her cool. I have not had a ton of time to read but from what I have read, the COBS can be driven just from the 110v power outlet? So youd be looking at probably another $200 for parts (if no drivers are needed). About $400-$500 for a 600w ACTUAL LED fixture which is really not bad at all.

My question then would be what to do about spectrums? With the COBs, they seem to only sell 2700k, 3500k, 4000k, 5000k (I think??). Do you add a few LED singles just for difference spectrums?

Ill definitely need to do some more reading this evening/week when I have time. Seems pretty interested although a little daunting at first as is anything. Half tempted to spend $600 on a dual 400w HPS setup to run for the next year while I experiment with LED tech.

http://www.alibaba.com/product-detail/CREE-CXA3070-Original-CREE-COB-LED_60022188146.html you may want to check these out, who knows? 10 min. order

 

[doz]

Ok, thanks for the info. Ill definitely get some more research done to better understand how they work. Id never order from Alibaba.com either.

As far as spectrums, just using red/blue is enough? No need for all these special spectrums that LED manufacturers claim? I know with the EYE HPS, even those provide small amounts of other spectrums, with the majority being yellow/red.

As far as 400w HPS, that cost is for 2x Solistek 600w Digital Ballast, 2x Eye Super HPS bulbs, and 1 reflector hood (I already have a hood). Id be cutting my wattage down by 200, but believe Id do just as well if not better w/ the proper coverage instead of 1 light in the center.

 

[bicit]

Ok, thanks for the info. Ill definitely get some more research done to better understand how they work. Id never order from Alibaba.com either.

As far as spectrums, just using red/blue is enough? No need for all these special spectrums that LED manufacturers claim? I know with the EYE HPS, even those provide small amounts of other spectrums, with the majority being yellow/red.

As far as 400w HPS, that cost is for 2x Solistek 600w Digital Ballast, 2x Eye Super HPS bulbs, and 1 reflector hood (I already have a hood). Id be cutting my wattage down by 200, but believe Id do just as well if not better w/ the proper coverage instead of 1 light in the center.

If you already have a hood, you could get two philips 315 CMH bulbs/ballast for ~$200 each from advanced tech lighting. They're a great company. More options to consider

 

[doz]

Also, where are people getting your housings from or are you building them? Like I said, I have access to smooth aluminum and can build a small fixture if needed. Just wondering if they sell them though, and if so which ones are quality?

Ill probably end up running my 1000w if I go the LED build side till its completed. Ill be flowering again here in the next month so I might just run with it for 1 more cycle.