Advice buying led bulbs

[Smokenpassout]

I was going to suggest led bulbs to a friend. Just bulbs not going to build a panel. My question is do the same rules as cfl apply? Should they use more 5000-6500k led bulbs to veg, then use more 2700k led bulbs to flower. Would cree be a good brand? What about distance on led bulbs? They dont need to be as close as 1-3" right. Should you get 40 or 60 watt or higher?

 

[ISK]

I was going to suggest led bulbs to a friend. Just bulbs not going to build a panel. My question is do the same rules as cfl apply? Should they use more 5000-6500k led bulbs to veg, then use more 2700k led bulbs to flower. Would cree be a good brand? What about distance on led bulbs? They dont need to be as close as 1-3" right. Should you get 40 or 60 watt or higher?

I just recently bought five 60 watt LED bulbs, both Phillips and Cree, three 5000K and two 2700K (they don't seem to make 6500k)

I choose 60 watts because they provide more lumens per dollar spent (60 watt has 800 lumens, and only 450 for the 40 watts)

I have no experience with LED, so I'm just starting to germ some bag seeds to start playing

I'm going to use them as I would CFL's, but I'm not sure about how far to keep them from the plants (I keep CFL's about 1 or 2 inches away)

For now, I'm adding a single 6500K CFL mostly for the heat, as the room is unheated and is cool, also for the additional spectrum can't hurt but is unlikely required

 

[churchhaze]

Those LED bulbs are only ~80lm/W. You'd likely do a lot better going with HPS unless you're interested in DIY with cree or vero cobs.

 

[az2000]

Those LED bulbs are only ~80lm/W. You'd likely do a lot better going with HPS unless you're interested in DIY with cree or vero cobs.

Rub the rubber film off the Cree lightbulbs and they're around 90lm/w. Remove the glass diffusion globe[1] and they're closer to 100lm/w.

I've been using them and they work fantastic. I just flowered a large plant under them at a remarkably low watts per sq ft and had great results. I did only 20-22w/sq ft. It would have been more appropriate at 30-35. But, the results were very good.

I like them. They're a bit of work spreading many light sources around a plant. I haven't grown with HPS because it's too hot for my climate. These bulbs run cool. No fans. I like the ease of going to the local store to get them. Not sure how an HPS would perform at 20-22w/sq ft. (I don't think it would be that good because I usually hear people using 50-60w/sq ft.).

To @Smokenpassout : Follow the link above where you'll find two more links to details about how I use LED "lightbulbs." You'll get some ideas about how to mount/position them.

[1] Electrocution hazard. Use a GFI outlet to minimize the risk.

 

[bicit]

The cree bulbs actually seem to work pretty well. I'd say t5s would be a better fit though. I guess it depends on your space requirement but those cree bulbs aren't cheap.

There was someone else on here who made a pretty big light using the cree bulbs from home depot. They also had good results.

I guess what bothers me about using them is that they emit light in all directions. Im betting the cree flood lights would work much better.

 

[az2000]

The cree bulbs actually seem to work pretty well. I'd say t5s would be a better fit though. I guess it depends on your space requirement but those cree bulbs aren't cheap.

There was someone else on here who made a pretty big light using the cree bulbs from home depot. They also had good results.

I guess what bothers me about using them is that they emit light in all directions. Im betting the cree flood lights would work much better.

CaptainMorgan had a thread using them. There is a link to that thread in my write-up which was inspired by his thread.

I started growing with T5HO. I felt they were hotter and didn't penetrate as well. New T5HO bulbs are 92lm/watt (from memory, maybe 96. I forget). They fall off quickly after usage due to lumen maintenance.

I don't know why lumens/watt doesn't translate perfectly into heat. I'd think watts not going to lumens would be heat. But, from my experience, CFL and T5HO run remarkably hotter than the Cree bulbs which are around 90lm/w with the rubber coating removed, and 100lm/w with the glass globe removed (use a GFI circuit to power them this way to avoid electrocution).

What I mean is: I can leave a plant touching a 9.5w (60w equiv) Cree bulb for an entire day and it won't burn. If I did that with a 15w (60w equiv) CFL the leaf would be burnt in 30-60 minutes. That's far more heat than the 30% lower efficiency. I don't know if it's generating more infrared. But, it doesn't correlate watt for watt the way I'd expect it to.

Regarding omnidirectional. T5HO and CFL are omni too. Their efficiency is reduced when placed in a reflector. I.e., reflecting the light causes less light to be emitted (even if it's more light reaching the desired location).

That's why Cree's PAR38s and BR19 (60w indoor flood) appear to be less efficient based upon the packaging's numbers. An onmi light like the Cree 19 would have similar numbers if is lumens were measured with the omni in a reflector.

The one thing that may make PAR38s more efficient is that they may be made with XTE diodes. The onmidirectional bulbs started with XTE, then went to XBD which is a step down. They're not bad, but not as good as XTE. I'm not sure if this was permanent or not because I'm still finding XTE models still in stock at Home Depot. (I don't spend much time differentiating because I don't think it matters much.). I don't know what the PAR38s use, but they could be XTE still.

I didn't feel T5HO penetrated as well. Also, being a flat panel it was harder to contour the lights to the canopy. The panel also blocked air circulation.

I think the big win with the LED lightbulbs is providing light from multiple angles. Bathing the plant with softer light creates more efficiency (IMO) than you can get from discriminating between diodes. In a 10x10 grow room this style of lighting would be too labor intensive. But, in a 4x4 tent it's doable. It's worth experimenting with. The worst thing that can happen is someone will end up with some spare LED bulbs they can use as supplemental lighting. That's always useful.

 

[Smokenpassout]

How much distance do you want with 60 watt cree leds? I know led panels to close can bleach. So a 60 watt cree is 60 actual led watts?

 

[TubePot]

FYI, the Cree 100 watt runs cooler than the 60 watt Cree....

 

[az2000]

How much distance do you want with 60 watt cree leds? I know led panels to close can bleach. So a 60 watt cree is 60 actual led watts?

A 60w Cree is 9.5w actual. You really don't need any distance. The leaves can touch the glass diffusion globe for hours (maybe a day or two) without burning. I haven't gotten any bleaching. I run them inside a clamp-on reflector (I adapt the reflectors to better mounting devices). No burn, no bleach.

I think the PAR38 spots need 12-14" or so or they'll bleach. When I flowered under five of them the plant stretched and got closer without me realizing it. I felt like the top leaves bleached. I can't be certain. I was having some issues reducing N too soon. And, I wasn't sure if it was an mg def which I can get in early flower. But, my opinion at the time was that they bleached.

For the PAR38 spots (and floods), I "focus" them. Shine them on a wall and see where you get the best spread and uniform brightness. For the spots, that's about 16", above the point of bleaching (if I did experience bleach).

I didn't get to try the floods as top lighting yet. They can be kept closer, maybe 6". They have a focus too. I forget what it is, but it's easy to shine one on a wall and see what looks best. It was about 9"

I haven't used the 75 and 100w equiv bulbs. I used the BR19 (60w equiv flood). I prefer the 60w equiv omnidirectional in a clamp-on reflector. The BR19 is good if space is tight. I think it's a little softer and not as penetrating as the same 9.5w onmi in a reflector.

 

[Smokenpassout]

The 60 watt omni directional crees would be a step on those 23 watt cfls right. Brighter? I am sure they dont need to be near 1-3" proximity like cfl. So maybe 4-6 could flower?

 

[az2000]

The 60 watt omni directional crees would be a step on those 23 watt cfls right. Brighter? I am sure they dont need to be near 1-3" proximity like cfl. So maybe 4-6 could flower?

If you want fewer lights, more light from fewer sources, I would use the Cree PAR38 spots or floods. They're 18w actual. It's like putting a 100w equiv (18w actual) Cree A21 omnidirectional in a clamp-on reflector.

If you have reflectors already, the A21 would be fine. The price difference between the PAR38 and A21 isn't substantial. I like the simplicity of the PAR38 and I figure the reflector engineered for that light is better than a generic clamp-on reflector. (The A21 doesn't fit the small clamp-on reflector well. It stands taller than that reflector. The next size up clamp-on is a little large for it.). The PAR38 is 3000k, not 2800k like the A21. It flowered well for me using A19s for sidelight.

You asked if the 60w (9.5w) A19 is a step up from the 100w (23w) CFL. I don't know how to compare these things. The 100w (18w) A21 produces the same lumens as a 100w (23w) CFL using 22% less energy. So, on its face, no, the 60w (9.5w) A19 isn't a step up. But:

1. Remove the gummy protective film from the glass globe and lumens rise ~10%.
2. The heat difference seems much higher than you'd expect from a 30% increase in energy consumption (from 18w A21 to 23w CFL). Perhaps the spectrum of the Cree is better than the CFL? Maybe the CFL produces more lumens in the infrared range?).
3. Despite the specs on the bulb packaging, the lumens per watt of CFL is around 65. The lumens per watt of XBD is around 100 (which you can get closer to by removing the glass globe, but use a GFI outlet to protect against electrocution. I don't do remove the glass.).


From my perception, I think the 75w (13.5w actual) A19 would be equivalent to (or a step up from) a 23w CFL. But, the lightbulb packaging doesn't support that view.

 

[bicit]

I know you said you weren't interested in DIY LED work. But the amount of money you're looking at spending on cree bulbs you could get a few vero 10's and pre-drilled heatsinks. Plus most of the labor would be the same as rigging up a dozen bulbs.

Heatsink link
driver
[/url=http://www.digikey.com/product-search/en?s=44063&pv1614=311&pv1615=1&FV=fff40008,fff80354&k=vero+10&mnonly=0&newproducts=0&ColumnSort=0&page=1&quantity=0&ptm=0&fid=0&pageSize=25]Vero 10 emitter[/url]
Pico EZ-mate(solderless connector)

Add a waggo 222 connector, a pair of M2.5 screws, and some thermal paste. You'll have an awesome passively cooled 7w light that's a touch about 40% efficient.

 

[az2000]

I know you said you weren't interested in DIY LED work. But the amount of money you're looking at spending on cree bulbs you could get a few vero 10's

What distance is required for that? I agree DIY COBs are more efficient than XBD/XTE diodes (used in the Cree lightbulbs). But, I believe distributing light among many sources creates its own efficiency. From what I've read in the COB threads, it sounds like they're typically run as more concentrated light. For more distributed lighting they have to be run at lower wattage, which gets more expensive in parts and labor.

Not trying to say "lightbulbs" are better. Just that there seems to be tradeoffs. I personally get nervous about something that runs so hot it can burn holes in floors. I like the fanless simplicity of these lightbulbs and their ubiquity. CFLs have that benefit too. And, if the efficiency difference was as great as it would be between COB and CFL, I wouldn't waste my time on CFL. But, the efficiency of these lightbulbs is up there enough that, with the theorized efficiency gains of multiple light angles, it seems to me like it's very close to a single high-powered COB.

 

[bicit]

6-12" hanging height is ideal. With the driver I linked it will run cooler than the cree bulbs and significantly cooler than a 23w cfl. Hottest part of the heatsink I was able to test was 86*f. Ill have to go back and look at my notes for the temp on the cree bulb.

I tested the cree bulb in a vertical reflector with no globe against a 4000k vero ten in the above configuration with a 90* lens. The lens reflected a lot of light back through the heatsink. But despite this I noticed no difference in performance between the two. But the vero is only 7 w

I need to rerun the test without the lens on the vero.

You don't need as many vero10's, and labor involves an Allen wrench lol. Not exactly rocket surgery tbh

 

[captainmorgan]

The Cree bulbs are also a great temporary light that are also cheap. When you upgrade to something better just use them around the house.