Quick question on LED panels

Reduced cost did occur to me. I guess there are several reasons a small unit would be more economical. Thing is, looking at the par footprint on the a51 website it looks like a HPS par footprint. If someone made a LED unit that gave an even 800-1000 par reading across the entire footprint I would be much more enthusiastic about spending the xtra money on them. I mean, in theory it would be superior right? No hotspot in the middle and the unit could be placed 6" from the plants, unless I'm missing something.

That Neosol is looking pretty good. I like the footprint and the open construction -w- passive cooling would allow multiple units to be placed side by side. I'm not understanding the numbers though. Does it actually run at 300 watts or is that the LED max rating? I'm going to do some research on it, see what I can find.

Any further comments on this topic would be appreciated.
 
I have the Apache at600 over a 4x4. @36" I get a 1000µmols center and 600 on the outer plants. That's better than 1000w hps. I'm about to flip in a few days
 
I have the Apache at600 over a 4x4. @36" I get a 1000µmols center and 600 on the outer plants. That's better than 1000w hps. I'm about to flip in a few days

That's a very even spread, but it's still being used 3 feet above the plants. The par 12" directly under it would probably be +2000. The par 12" under the center of the new A51 model is over 1500.

I like the NeoSol design. It seems it's been out more than a year now, but there's not much info about it being used. I did find http://www.hydropatio.com/products/neosol-ns-fixture-led-grow-light some pictures at a retail site. It's got a nice looking footprint pic showing visible light coverage, but again it shows it being used 2-3 feet above the plants in another pic. Still, that footprint looks sweet for 300 watts... to bad there are no par values, or grow logs that I could find. I'm lighting a 3x6' area so I would use 2. It would be $1000 more than choosing 3 of the A51 units which I really like atm. The Neosol is fanless but I don't see how that justifies the premium.

Anyone ever use a NeoSol 300w?
 
3 a51's all day for that space. sgs160's or xgs190's. Like you say there are very few actual uses of neosols. As where you have average of 5 per page here for a51 of good to amazing results to validate what they have going on.

Not to mention the extra cost for un validated results. And illuimtex is supposed to be releasing new ones soon. Basically an upgrade. So you don't want to blow your load on them yet if you choose them.
 
That's a very even spread, but it's still being used 3 feet above the plants. The par 12" directly under it would probably be +2000. The par 12" under the center of the new A51 model is over 1500.
Anyone ever use a NeoSol 300w?

You are always going to get higher PAR readings at the center of a panel. Spread is going to be more even at 24 inches than at 6. Plants aren't going to efficiently use 2000 umoles. I do like how the xgs190's can be run as 130s.
 
You are always going to get higher PAR readings at the center of a panel.

I think you're right but I think spacing things out could possibly improve the difference. I'm looking into DIY. Not sure it will save me much but it would give me exactly what I have in mind.

I like the 190 too. Running at 130 is perfect for growth and 190 would support a good sized bush.
 
I'm looking into DIY. Not sure it will save me much but it would give me exactly what I have in mind.


It cost me about $4/watt for parts but I put a lot of emphasis on efficiency and passive cooling and ended up with about 41% efficient overall (average radiometric efficiency, at operating temperature). Since then there gave been some game changers, the Vero COB and the XML2. If you used just the Vero 18 3000K at 350mA the lamp would be 37.8% efficient and slightly cheaper although I don't think the spectrum would be ideal to get HPS-like nugs. It needs more deep red to get the big dense nugs and more blue to get the frosting (improvement over HPS).


If I were to design the perfect DiY I would stick with the same spectrum I have been using, but swap out the XTE WW with XML2 WW T5 bin and use less of them. Cost would be slightly higher but efficiency would rise to 43%. If we could get our hands on the Oslon SSL deep red 4T bin, efficiency would rise to 46%. Junction temps are about 50c so the lamps will maintain their output indefinitely so the extra cost over hps MAY actually get returned over time in electrical savings or even increased yields.
 
Apache might be one of the best commercial LED lights on the market? It cost $4-$6/watt for the AT series. I am skeptical about its design and claims along with every other commercial panel I have studied. The main reason is because of efficiency. 600HPS is 35-40% efficient when brand new and a pretty decent flowering spectrum at 2000K. 600HPS cost $200 for bulb ballast socket and reflector. So that is the competition if you are building a large flowering LED.


-There is no mention of the brand or bin of LED they are using. This is critical. Cheapo LEDs running hot could be as low as 15-20% efficient. When they don't specify bin, that is either because they don't know or because they know better ;) You should assume the worst here.


-There is no 660nm in the spectrum, love it or hate it is it better to spread out your red as much as possible.


-Way too much blue. I have gone as high as 20% during flower but they are 35-40%.


-Compact design requires active cooling wasting power.


-High input power, how efficient are the drivers, how much power are the fan using, how much heat am I adding to my room? The AT120 draws 180 watts. Ideally that number should be 132. Either they are driving the LEDs harder than 350mA, the drivers are very inefficient or the fans are robbing tons of power. If an AT120 draws 180 watts, does the AT600 draw 900 watts? If so, there is no significant power savings over the 1100watts of HPS. Plus with the HPS we can keep the 100w of ballast heat out of the room. If these LEDs are running at only 20% efficient as I suspect they are, these lamps actually put more heat in your grow room than HPS and charge you 15X the price for it.


-Shining through lenses or acrylic covering. Lenses and coverings rob 10-15% of photons. This is required in order to have a compact marketable design, but better to be bare and spread out.


-14 degree lenses require tall ceilings and waste vertical space. If you keep them too tight on the canopy it will just cover a smaller area and potentially bleach them with hot spots and poor color mixing.


I'm not hating on Apache just trying to be objective about commercial LED capabilities. It would be very hard to commercialize the KNNA DiY LED design and it is very time consuming to build. But if you want the best you got to do it yourself.
 
Thanks for the comments Supra.

I've been playing around with cone angles and watts. I need to work up a list of options and real figures but the basic gist is to place a 115 degree higher wattage emitter in the center and 1 or 2 tighter (50 or 90 degree) cones on the outside. With the right wattage differential I think I might be able to achieve higher par reading in the middle/outer edge than in the center. Is this something you've considered?

OTOH since I'll have reflective material on the walls I could work with 2 rows @ 115 degrees, get a pretty good spread across the center and rely on reflection to boost the outer edges. When I start looking at the various output and angles available and try to consider the additive values it gets complicated trying to work out an even par spread. I also suspect the center of any angle is going to run brighter, further complicating things.

Ideally I would like to purchase some test subjects and observe the footprints and additive values at various heights inside the grow space with my eyes but am not sure how deep I want to go in that direction.
 
Apache might be one of the best commercial LED lights on the market? It cost $4-$6/watt for the AT series. I am skeptical about its design and claims along with every other commercial panel I have studied. The main reason is because of efficiency. 600HPS is 35-40% efficient when brand new and a pretty decent flowering spectrum at 2000K. 600HPS cost $200 for bulb ballast socket and reflector. So that is the competition if you are building a large flowering LED.


-There is no mention of the brand or bin of LED they are using. This is critical. Cheapo LEDs running hot could be as low as 15-20% efficient. When they don't specify bin, that is either because they don't know or because they know better ;) You should assume the worst here.


-There is no 660nm in the spectrum, love it or hate it is it better to spread out your red as much as possible.


-Way too much blue. I have gone as high as 20% during flower but they are 35-40%.


-Compact design requires active cooling wasting power.


-High input power, how efficient are the drivers, how much power are the fan using, how much heat am I adding to my room? The AT120 draws 180 watts. Ideally that number should be 132. Either they are driving the LEDs harder than 350mA, the drivers are very inefficient or the fans are robbing tons of power. If an AT120 draws 180 watts, does the AT600 draw 900 watts? If so, there is no significant power savings over the 1100watts of HPS. Plus with the HPS we can keep the 100w of ballast heat out of the room. If these LEDs are running at only 20% efficient as I suspect they are, these lamps actually put more heat in your grow room than HPS and charge you 15X the price for it.


-Shining through lenses or acrylic covering. Lenses and coverings rob 10-15% of photons. This is required in order to have a compact marketable design, but better to be bare and spread out.


-14 degree lenses require tall ceilings and waste vertical space. If you keep them too tight on the canopy it will just cover a smaller area and potentially bleach them with hot spots and poor color mixing.


I'm not hating on Apache just trying to be objective about commercial LED capabilities. It would be very hard to commercialize the KNNA DiY LED design and it is very time consuming to build. But if you want the best you got to do it yourself.

They are pricey no doubt. But the parts they use are the reason. As well as it is actually made in america, not just designed. They deal straight with nichia and have access to top bins. I don't have the bin numbers but I have the performance numbers in person to show for it.

Old at was 158w. Not sure on the chip model but they were a nichia...and nichia is a big reason for the cost, in the new units as well.

AT200 is 196w
At600 is 672w
Both are under 30C while operating. I will say they changed fans a year ago or so and it keeps them cool and they don't fail, but its way louder than the original.

Both the new models are using nichia219b. and I believe its around 4200K. There is 660nm being emitted, just not directly an they believe it's enough. The 219b reds are more intense than the old at120's chip so it is more red dominate now too. The warmer white drops the blue down. They have a plant physiologist and they are totally confident in their spectrum. But they are also not MJ growers primarily. That's why the R2 spectrum was never put on the site, because greenhouses only want the R1, as where the mj is the reason for the R2... whether they'll say it or not, that's what the R2 was made for.

Drivers were and still are meanwell plc's...they have been testing a few newer ones. HLG was one but not sure what the thoughts were on it.

The 14 lens is actually 28* in field of view terms. That's still tight, but not as bad as it seems. They have 50 refractions too, which is 100* in field of view. At 24" above it still out peforms the 1000hps over a 4x4. The 50's can be dropped lower and performance can be it the same. I tell them all the time 30refractions(60* FOV) would be perfect. I think they are moving on in design soon.

The design is there limit right now. I think they could get the same performance as the at600 is now with about 530w and a redesign. Tech has improved and there design isn't fully taking advantage of the techs capabilities. They drive them soft so efficiency and longevity are great but they can easily handle more and space ten out to use less material and in turn bring cost down. But as far as commercial panels go, they perform like they are supposed.


I have always heard about KNNA, but wan't till they other day that I finally was shown a thread of his. What a genius and way ahead of the times. Basically everything I have learned through my trials...he already did and knew. And I would have save a bunch of time.

What's his DIY design, you got a link?
 
Ive done small grows inside mylar boxes with bare emitters and that worked out very well. The LEDs were 6 inches or less from the canopy and will reach budsites that are about 12" from the top of the canopy (if they are not shaded). But for large grows, I discovered that even with bare 600HPS and reflective walls, shaded bud sites tend to be sucker buds (lighter green, leafier and less dense). In other words, I was misunderstanding what penetration really meant. It might be better described as "reach".

It turns out that having a long reach comes at the cost of efficiency (lenses, reflectors and bright hotspots (HPS). So rather than to try and extend reach, I just try to manage my canopy better. I pluck out the sucker buds and underbrush during late veg and early flower. When I get colas that jump out of the canopy, I bend and crease them depending on where it is. Some plants get spread open with a wire of two. I adjust the height of each LED module every few days but I don't put too much time or care into any of it, just bang it out real quick because I don't want to create additional work.

All that said, if you need tight angles for your particular setup, you could look at the Oslonn SSL80s. I have a few of them in deep red and so far no complaints. Another thing that affects reach is of course how hard you drive them. Most of my LEDs are dissipating 1.5 to 2 watts. But a Vero18 @ 350mA would be dissipating 10 watts in a relatively small area (~1 cm2 I think) so that would reach lower budsites better for example.

But when it comes down to it the plants are not terribly picky. I have bleached and burnt a few shoots from time to time that shot up into the light. It happens but regardless, the improvement over HPS has been great. Less than half the power used for the same yield. Nugs are denser, icier and stinkier. Whatever design you go with, itll get the job done.

Bunch of OGs last run
DSC06692a.jpg
 
That OG looks killer. I see your distance there.

I have not gotten any bleaching since going to white panels.

I like the "reach" thought. The quanta is still there and not escaping if they are ran close. Plus if the canopy is wide and full, there isn't a need to reach under too far down.

How many of your panels do you use over a certain space?
 
Ya I am in love with the OGs especially under LED. In that pic the LEDs were pulled up from the canopy just so the camera could get the buds. This is probably a better pic to show how tight I keep them. In this pic they just went into flower and the underbrush will get thinned out. As the plants continue to grow the first few weeks I move the buckets apart so they can spread out into the space. After a few weeks I kicked on a bare 600HPS right in the middle. (The pic above were LED nugs only)
IMG_9476a.jpg

color corrected you can see better
IMG_9481a.jpg

Currently there are (5) 50 watt panels and their job is to replace a 600 HPS. They successfully did the job with 4 panels, but only just barely. I have to finish up soldering the sixth panel and then I should have increased the yield beyond what the 600 could do. Once this canopy spread out it is 4'X4'. Sometimes they get as wide as 6'X4' depends on the batch.

Same ladies at 4 weeks. You can't see the HPS, it is 12-18" above the canopy bare bulb.
IMG_9566a 4 weeks.jpg
In the foreground are RP Kandy Kush. The Ace of Spades is on the left and SSSHD (Shire cut) are in the middle and background. None of these varieties are big yielders excpet maybe the Shire. Normally when I find a big yielder I am disappointed with the quality. The Ace of Spades is the fastest it just flopped over had to tie it up.

Here is what a pair of brand new bare 600 HPS can do with decent yielding varieties. The Shire are in the background. 1.5LB dry, tough to beat.
DSC06100a.jpg
 
RE Apache. Sounds like they made some decent upgrades from 6500 to 4200 and more red. The Nichia 219 is decent but of course the bins range from 100 to 135 lumens/watt, so that is important but it sure is a heck of a lot better than "unknown chinese white LED" lol. At 100 lumens/watt the neutral whites are 30% efficient and at 135 they are 40%. Those are the numbers for the 70 CRi but Nichia tends toward High CRi. If Apache went that route the number would be much lower but I didn't check.

The youtube I saw was the AT120 and it was drawing 180watts. They didn't let it warm up though. Meanwell PLC are nice, they claim 89% but I am not sure the consequences of turning them down from 1.3A to .35A.

I would be very interested to know the heatsink temp between the fins, inside the case, when the unit is running. The Nichia NW does not get badly punished by excess heat but reds typically do. That would also depend on how they were mounted to the heatsink. But since they are running at 350mA or so there is not as much opportunity for very high junction temps.

RE KNNA
Yes it was awesome having him working on these lamps with us. The basics of his design are
-Passive cooling, 135cm2/watt (now I use 110cm2/watt because radiometric efficiency has improved).
-15-20% blue for flowering 25-30% blue for vegging
-AC-DC Constant current drivers
-Bare leds, no lenses or reflectors, keep modules in tight on the canopy
-Spread the light onto the canopy as evenly as possible
-Run the LEDs as soft as you can afford to, and within the constraints of reach.
-You can vegetate with just red/blue but some varieties appreciate the addition of white for flowering (back in those days we had to use cool whites)
-You can grow with just blue and 630nm reds but not with just 660nm reds
-Top bins for radiometric efficiency was critical in those days
-We used copper traces and kapton tape to hand mount the LEDs to the heatsink

I still have some vegging LEDs in use with that design
IMG_9572a.jpg
 
I met him on Gardenscure in 2008. This was the original thread. He was on IC also. I think he tried RIU but he got a lot of flame from the HPS guys, even though he is a big fan of HPS lol.
 
Why do the good panel makers bunch the LEDs up in a small area?

Who cares?
I am not buying into LED hype. Let me know when LEDs will match HPS with overall efficiency cost/ price included. I'll buy into it then. A'int happening very soon.

Commercial growers know that and they still stick with HPS. I am not going to try to outsmart those guys.
 
Any commercial growers that aren't using LED for vegging are seriously missing out.

But regarding flowering, with electricity at 12-15 cents per KWH you are correct, the up front cost difference is extreme. Maybe obummr will slap a big fat tax on electricity and change all that lol.

For commercial grows LEDs can also provide stealth since we don't need huge fans to cool the HPS hoods. I love having a quiet grow, using ceiling fans for circulation and max fans for scrubbers set on LOW. There is also lumen maintenance to consider. HPS output fades over the months but LEDs running cool will still be at 95% after 10 years. When you add up the electricity savings and the cost of replacement HPS bulbs it is not such a false economy. If you run a commercial operation the LED pays for itself in 1 month. And finally, if you want a leg up on the competition, LED nugs are danker and have more bag appeal.
 
Back
Top