CXB3590 1500W

[SupraSPL]

That Alien looks great!

It is a stinker, coffee + OG cologne, good potency, bud density and plant structure, mold resistant, one of my favorites maybe top 10 out of 200.

I have bought 2 heatsinks 5.88" x 19" so I can put 2 cobs on each one. My question is regarding the spacing of the cobs on each heat sink.

I noticed you installed the cobs closer to the corners on one more than on the other. Does this help,hurt, or make any difference at all in temp. control. What would you recommend the distance be on my set up from cob to cob and the edge of the heat sink?

Yes I think when push comes to shove uniformity of PPFD is more important. Poor temp distribution in your heatsink might cost you only 1% total output but poor uniformity could potentially cost much more



You can have it both ways but it requires playing around with heatsink profiles. This is a 2' X 6' with the heatsinks also to scale. Based on this I would suggest mounting the COB close to the edge or even slightly diagonally if necessary.

 

[ttystikk]

It's a balance between optimizing for light distribution and temperature distribution. Some (SDS) would argue that optimal temperature distribution at the expense of suboptimal light distribution is better while others (me and Supra) optimize more for light distribution at the expense of suboptimal temperature distribution.

I just ran the numbers for this setup to cover four ft² and got a PPfD of 1236. First, is this correct? Second, is it way too much? A dimming pot to cut output by half would still give 618 PPfD... which seemed like plenty for veg and hardening the plants off to the higher irradiance.

I'm seriously considering this combination because of relative efficiency and the fact that it uses four COBs for four ft². Six of these modules would cover my 4' x 6' trellis evenly, arranged like a sixpack.

Would it be better to go for a lower irradiance? I have plans for dealing with the heat generated, I'm just worried about cooking them with too much of a good thing.

@SupraSPL and anyone else, I'd appreciate whatever input and insight you can offer.

 

[Growmau5]

Your setups always inspire me. I'm planning my setup for the second half of my room and probably going the same direction as you: CXB3590 3500K CD bin run at 700mA. My question is about temperatures of the lens and metal collar around the lens. Hypothetically, if a person were to mount a 3590@700mA with this same lens setup to a piece of wood, would it be hot enough to catch on fire? just curious. wondering how hot these areas are in the attached pic.

 

[SupraSPL]

I just ran the numbers for this setup to cover four ft² and got a PPfD of 1236. First, is this correct? Second, is it way too much? A dimming pot to cut output by half would still give 618 PPfD... which seemed like plenty for veg and hardening the plants off to the higher irradiance.

I'm seriously considering this combination because of relative efficiency and the fact that it uses four COBs for four ft². Six of these modules would cover my 4' x 6' trellis evenly, arranged like a sixpack.

Would it be better to go for a lower irradiance? I have plans for dealing with the heat generated, I'm just worried about cooking them with too much of a good thing.

@SupraSPL and anyone else, I'd appreciate whatever input and insight you can offer.

I came up with 1244 also, very high intensity. I have not experimented with numbers that high but with the right varieties, if you can keep up with feeding, temp and CO2 you might be able to get a record harvest from a given space. There would be some sacrifice of efficiency because of the law of diminishing return photosynthesis is much more efficient at lower intensities, but it might maximize space efficiency.

There is also the possibility that intensity that high could affect potency and flavor negatively, something to be on the lookout for if you are focused more on medicine than weight. Or it may be more dependent simply on keeping bud temperature down in the final weeks?

 

[SupraSPL]

Your setups always inspire me. I'm planning my setup for the second half of my room and probably going the same direction as you: CXB3590 3500K CD bin run at 700mA. My question is about temperatures of the lens and metal collar around the lens. Hypothetically, if a person were to mount a 3590@700mA with this same lens setup to a piece of wood, would it be hot enough to catch on fire? just curious. wondering how hot these areas are in the attached pic.

View attachment 3485051

A valid concern, the extremes of intensity occur just in front of the LES. That CXB3590 pictured was tested for temp droop, driven at 2.55A (100W). So the unit was left running for at least 30 minutes at 100W and there were no issues with the lens frame temp, although I did not measure it and now I am curious. Since you would be running at 25W ea you are well within safe limits.

 

[Growmau5]

my question stems from an idea I had about building a new fixture. I want to build a massive box. 7 feet long, 40" wide, 6 or 7" thick. 32 cxb3590s on passive heatsinks would be encased in the box with just the lens protruding from 32 individual holes. This would put the side of the lens near the metal brackets of the lens in direct contact with plywood. The whole thing would be air cooled just like a massive HPS light. Building a single fixture on this scale out of metal would be tedious for me, and cost a ton. If i could build it out of wood with out it catching on fire, it would be so much easier. The efficiency of the 3590@ 700mA plus removing the passive heat from the heatsinks completely out of the room would be the ultimate, for me. Just putting feelers out right now as I make sketches.

 

[ttystikk]

I came up with 1244 also, very high intensity. I have not experimented with numbers that high but with the right varieties, if you can keep up with feeding, temp and CO2 you might be able to get a record harvest from a given space. There would be some sacrifice of efficiency because of the law of diminishing return photosynthesis is much more efficient at lower intensities, but it might maximize space efficiency.

There is also the possibility that intensity that high could affect potency and flavor negatively, something to be on the lookout for if you are focused more on medicine than weight. Or it may be more dependent simply on keeping bud temperature down in the final weeks?

What's the optimal amount of irradiance? 1000PPfD?

 

[SupraSPL]

What's the optimal amount of irradiance? 1000PPfD?

That is a good question and not all PPFD is created equal. For example if the PPFD is more uniformly distributed you could use less to get the same job done. Another example, high CRi PPFD may be more effective than low CRi PPFD. Warm/red tinted PPFD should be more effective than blueish tinted PPFD. If you are using a shelf system with short ladies you might be able to run a lower PPFD. Soome varieties may respond differently to high or low intensities.

All that said, I use 750 PPFD average and have been happy with that intensity, nugs just the size I prefer. Captain Morgan has been having success in the 500-600 range if I recall.

For reference, a brand new 1000W DE HPS in a 4X4 is about 950 PPFD averaged. It is 40 CRi and relatively poor uniformity. A standard 600 HPS in a 3X3 is about 910 PPFD and 20 CRi.

 

[SupraSPL]

my question stems from an idea I had about building a new fixture. I want to build a massive box. 7 feet long, 40" wide, 6 or 7" thick. 32 cxb3590s on passive heatsinks would be encased in the box with just the lens protruding from 32 individual holes. This would put the side of the lens near the metal brackets of the lens in direct contact with plywood. The whole thing would be air cooled just like a massive HPS light. Building a single fixture on this scale out of metal would be tedious for me, and cost a ton. If i could build it out of wood with out it catching on fire, it would be so much easier. The efficiency of the 3590@ 700mA plus removing the passive heat from the heatsinks completely out of the room would be the ultimate, for me. Just putting feelers out right now as I make sketches.

I am confident it would not ignite the wood but it may be considered an inherently dangerous fixture by technical standards. Maybe you could wrap the closest parts of the wood with heavy duty aluminum foil. Some diyers use angle aluminum to strap heatsinks together. I like the idea of a large cooled box but on the other hand I have an unruly canopy and need lots of adjustabilty LOL

 

[MarWan]

"From an appearance standpoint, unprotected plywood should not be used when temperatures exceed 200° F (93° C). At temperatures above 200° F, plywood undergoes slow thermal decomposition that permanently reduces its strength. Between 70° F and 200° F strength loss is recovered when temperature is reduced. Between 70° F and 200° F, the need for design adjustment depends upon whether or not the plywood moisture content is reduced by the elevated temperature. Exposure to sustained temperatures higher than 200° F (93° C) will result in charring and weight loss. Using plywood in applications involving periodic exposure to temperatures from 200° F to 302° F (93° to 150° C) should be based on the amount of exposure and the amount of decomposition that can be tolerated without impairing the serviceability of the panel.

One example of plywood use in extreme conditions involves plywood pallets used in an annealing oven. Although the temperatures reach 350° F, the plywood performs well, despite the slight charring and discoloration."

http://www.performancepanels.com/thermal-properties

 

[ttystikk]

That is a good question and not all PPFD is created equal. For example if the PPFD is more uniformly distributed you could use less to get the same job done. Another example, high CRi PPFD may be more effective than low CRi PPFD. Warm/red tinted PPFD should be more effective than blueish tinted PPFD. If you are using a shelf system with short ladies you might be able to run a lower PPFD. Soome varieties may respond differently to high or low intensities.

All that said, I use 750 PPFD average and have been happy with that intensity, nugs just the size I prefer. Captain Morgan has been having success in the 500-600 range if I recall.

For reference, a brand new 1000W DE HPS in a 4X4 is about 950 PPFD averaged. It is 40 CRi and relatively poor uniformity. A standard 600 HPS in a 3X3 is about 910 PPFD and 20 CRi.

Good points, all... dammit- you are not making my job easier! LOL

Currently, I'm running 24-30W/ft² of 860W CDM lamps, CRI 92, color temp 4000 Kelvin. I'd like to see the math converting this lighting to PPfD so I can compare.

I run these lamps vertically and bare, so there are no lens or reflector losses. Four or five of them illuminate an oval/ellipsoid trellis rack of 144 ft². Losses from the top or bottom are reflected back in by panda film, I believe my overall lighting efficiency in getting photons onto leaves is well beyond 90%- at least 40% better than a typical flatlander facility where light is leaking all over the place.

My evil plan is to build a COB LED system for the same trellis system and I don't want to be limited by less than optimal irradiance.

Even at that 1240PPfD value, we're still talking 110W of dissipation/4 ft² = 27.5W/ft²... or is there something I'm missing?

 

[SupraSPL]

I do not have the QER data for the metal halide but being 4000K and high CRi the spectrum effectiveness should be similar to the 3500K 80 CRi Crees. Do you mean 24-30 PAR W/ft² or dissipation W? How much canopy surface area does each bulb illuminate?

 

[ttystikk]

I do not have the QER data for the metal halide but being 4000K and high CRi the spectrum effectiveness should be similar to the 3500K 80 CRi Crees. Do you mean 24-30 PAR W/ft² or dissipation W? How much canopy surface area does each bulb illuminate?

Metal halide is different from 860W CDM.

Link;
http://advancedtechlighting.com/cdmea860.htm

I mean that I'm running either 4 or 5 (depending on summer temperatures) lamps, so;
860W x 4 = 3440W/144 ft² = 24W/ft²
860W x 5 = 4300W/144 ft² = 30W/ft²

I was asking you to help me find PAR and PPfD... so that good comparisons can be made. I'm hoping some of lamp specs on that link will help?

The lamps hang vertically inside an oval shaped rack, so light from every bulb reaches every corner of the trellis.

The 56% efficient setup over 4 ft² seemed to run about the same watts per square foot as I'm running now; 110 dissipation watts / 4 = 27.5W/ft².

Trust me when I tell you that I have the heat part of things well in hand in the plans! It's the irradiance I'm concerned about.

 

[OneHitDone]

my question stems from an idea I had about building a new fixture. I want to build a massive box. 7 feet long, 40" wide, 6 or 7" thick. 32 cxb3590s on passive heatsinks would be encased in the box with just the lens protruding from 32 individual holes. This would put the side of the lens near the metal brackets of the lens in direct contact with plywood. The whole thing would be air cooled just like a massive HPS light. Building a single fixture on this scale out of metal would be tedious for me, and cost a ton. If i could build it out of wood with out it catching on fire, it would be so much easier. The efficiency of the 3590@ 700mA plus removing the passive heat from the heatsinks completely out of the room would be the ultimate, for me. Just putting feelers out right now as I make sketches.

Couldn't you just open up the clearance between the metal parts of the led assemblies and the wood?
Air cooled cob fixture is sounding very sexy

 

[SupraSPL]

Metal halide is different from 860W CDM.

Link;
http://advancedtechlighting.com/cdmea860.htm

I mean that I'm running either 4 or 5 (depending on summer temperatures) lamps, so;
860W x 4 = 3440W/144 ft² = 24W/ft²
860W x 5 = 4300W/144 ft² = 30W/ft²

I was asking you to help me find PAR and PPfD... so that good comparisons can be made. I'm hoping some of lamp specs on that link will help?

The lamps hang vertically inside an oval shaped rack, so light from every bulb reaches every corner of the trellis.

The 56% efficient setup over 4 ft² seemed to run about the same watts per square foot as I'm running now; 110 dissipation watts / 4 = 27.5W/ft².

Trust me when I tell you that I have the heat part of things well in hand in the plans! It's the irradiance I'm concerned about.

I have not made a good study of it, in what ways is ceramic metal halide different than standard metal halide, mostly just CRi?

OK so you are talking dissipation W when you say 24-30W/ft² but the COB setup is 196 dissipation W -> 110 PAR W -> in a 4X4 = 27.5 PAR W/ft².

So if we assume brand new CMH is 40% efficient and QER is 4.5:
27W * .4 = 10.8 * .9 scatter = 9.72 PAR W/ft² * 4.5 umol/s/W * 10.7 = 468 PPFD averaged.
27W * .4 = 10.8 * .95 scatter = 10.26 PAR W/ft² * 4.5 umol/s/W * 10.7 = 494 PPFD averaged.

So the COB setup at 1240 PPFD would be an extreme increase in intensity, almost double what I normally recommend as a fair target (750). That may actually cause some problems in terms of keeping the ladies fed and controlling bud temperature, even with good ambient temp control.

 

[ttystikk]

I have not made a good study of it, in what ways is ceramic metal halide different than standard metal halide, mostly just CRi?

OK so you are talking dissipation W when you say 24-30W/ft² but the COB setup is 196 dissipation W -> 110 PAR W -> in a 4X4 = 27.5 PAR W/ft².

So if we assume brand new CMH is 40% efficient and QER is 4.5:
27W * .4 = 10.8 * .9 scatter = 9.72 PAR W/ft² * 4.5 umol/s/W * 10.7 = 468 PPFD averaged.
27W * .4 = 10.8 * .95 scatter = 10.26 PAR W/ft² * 4.5 umol/s/W * 10.7 = 494 PPFD averaged.

So the COB setup at 1240 PPFD would be an extreme increase in intensity, almost double what I normally recommend as a fair target (750). That may actually cause some problems in terms of keeping the ladies fed and controlling bud temperature, even with good ambient temp control.

So what I'm getting from this is that I need to find a way to effectively light 6 ft² with that setup, rather than cramming it into 4 ft². If I can use 4 COBs to evenly illuminate a 2' x 3' rectangle, we're in business!

What's QER? How is 'scatter' determined? I was going to run the above for 24 and 30 dissipation watts/ft² to see where I am right now.

 

[Rahz]

my question stems from an idea I had about building a new fixture. I want to build a massive box. 7 feet long, 40" wide, 6 or 7" thick. 32 cxb3590s on passive heatsinks would be encased in the box with just the lens protruding from 32 individual holes. This would put the side of the lens near the metal brackets of the lens in direct contact with plywood. The whole thing would be air cooled just like a massive HPS light. Building a single fixture on this scale out of metal would be tedious for me, and cost a ton. If i could build it out of wood with out it catching on fire, it would be so much easier. The efficiency of the 3590@ 700mA plus removing the passive heat from the heatsinks completely out of the room would be the ultimate, for me. Just putting feelers out right now as I make sketches.

I know from experience that the 12 ounce radials work well passively with Vero 29 at .7 amps (39.5 Tj) with only tent vent fan. CXB3590 at .7 would have even better thermal performance without air cooled hood. Still, 32 of them, 21 lbs. of metal in the sinks alone. Suggest full passive on aluminum frame, split the lamp in 2 unless you want it to weigh a ton. My lamp has 23 emitters and is heavier than I would like.

 

[ttystikk]

I know from experience that the 12 ounce radials work well passively with Vero 29 at .7 amps (39.5 Tj) with only tent vent fan. CXB3590 at .7 would have even better thermal performance without air cooled hood. Still, 32 of them, 21 lbs. of metal in the sinks alone. Suggest full passive on aluminum frame, split the lamp in 2 unless you want it to weigh a ton. My lamp has 23 emitters and is heavier than I would like.

I have a plan to solve this problem... and even put the heat to good use.

 

[HydroDC]

Couldn't you just open up the clearance between the metal parts of the led assemblies and the wood?
Air cooled cob fixture is sounding very sexy

I think this is exactly right. A 1/6th" or maybe even a 32nd would answer well.

 

[SupraSPL]

So what I'm getting from this is that I need to find a way to effectively light 6 ft² with that setup, rather than cramming it into 4 ft². If I can use 4 COBs to evenly illuminate a 2' x 3' rectangle, we're in business!

What's QER? How is 'scatter' determined? I was going to run the above for 24 and 30 dissipation watts/ft² to see where I am right now.

Yes in a 2X3 it would be 915 PPFD averaged.

QER represents how many photons/s/PAR W that particular color temp/tint emits. You can derive that figure from analyzing the SPD curve. I prefer to measure it from 400-700nm because we are looking for PAR data, but you can measure it from 380-780nm as well.

When it comes to scatter I am just estimating. Any light that is hitting the wall will get scattered in all directions and it gets taxed on each bounce. So we cannot assume that 100% of the photons emitted from the bulb will reach the canopy even in a vert setup.