CMH / QB288 / QB120 / SB240 / VERO 29C comparison...

Photon Flinger said:

What is the primary reason for the upgrade? Reduce costs? Replace failing equipment? Increase yields?

Click to expand...

hold on... I think I answered this already... let me find it..
ah yes..

ganjamystic said:

I'm ready to send the blurples on their way though and upgrade the lighting to something more efficient (same amount of light for less W and with less heat, rather than more light for same amount of W and heat.. if that makes sense... I in other words, I don't need more light in the space, just less heat and a lower power bill).. and honestly I just want the light to be all warmish to neutralish white because it's so much easier on my eyes and I just want to be able to wear regular shades in there not green ones ...

Click to expand...

basically, I'd like to keep the yields the same (pretty well maxed out for the square footage without going vertical), but reduce electrical usage/cost and lower heat (further reducing electrical cost by cutting down on AC usage)... and I'm tired of wearing green shades...

Photon Flinger said:

You already have blurples, since you are considering DiY, would upgrading those units be a possibility? Or even cost recovery selling them used to help with the budget?

Click to expand...

yep, they're already sold actually... I'll send them on their way as soon as I have replacements... I still have some other even older ones I'm keeping to mess around with... I have a thread on modifying them for greater efficiency... but that is a super low priority side project, just for fun...

Photon Flinger said:

Do you have to do this all at once or in stages? Staggering out your build 1 tent at a time is something to think about.

Click to expand...

I'm upgrading the whole 7x7 space all at once.. I may start with a lower number of boards or cobs at first and add more later to boost efficiency, spread, etc..

Photon Flinger said:

First things first, you should get the PAR meter. $150 investment for the best tool (besides plant themselves but plants are a bit slower) to dial in your lighting. Take that PAR meter and measure your current setup so you know what is working and have a baseline to compare with.

Click to expand...

ok guys I get it, I need to get a PAR meter! lol

the blurples I'm currently using are inefficient so they pull high watts and generate lots of heat... 2 of them are 750w (actual draw) and one is 870w (it's actually two separate 435w lights but same difference)... in contrast, the qbs in the remaining quadrant only draw 350w total and are just as bright (according to my eyes and the plants haha)..

Oh yeah I thought I recognized you from somewhere. How is that project going? I have to say that the blurples once adjusted are great en masse for seedlings and early vegging. With cobs from seed they seem to be lazy.

Good on you for the PAR meter decision. You won't regret it.

I would suggest you start with a target power budget and build around it. For example, you are running around 2400w plus 350w for qbs if I got that correctly for 49ft2 or roughly 2 5'x5'. Since you are a good grower and plan to increase efficiency, I think a 25w/ft2 target would be desirable.

That puts you at 1250w total. Play the balancing game to figure out the most cost effective way you are comfortable with for drivers. Whatever money you have left is going into led and heat sinks. Once again, most cost effective way that works with the drivers you are comfortable with.

When you get more money buy more of whatever light source you chose and put them in parallel on the existing drivers for an efficiency bump resulting in a reduction of heat.

That help?

Photon Flinger said:

Oh yeah I thought I recognized you from somewhere. How is that project going?

Click to expand...

haha on hold for the moment.. my attempt to wire the 5 blurple modules in parallel to one driver didn't work out.. I have a feeling the driver might be bad, so I'll try again with another one. I'm fairly certain my wiring was right. you can check it out on that thread and let me know what you think.. when I tried to fire it up it would pulse/flash super bright about once per second, but wouldn't stay constantly lit (and the flashes looked way too bright), so I dunno what was going on there.. anyway, if you have any insight on that just post it up on that thread..

Photon Flinger said:

I have to say that the blurples once adjusted are great en masse for seedlings and early vegging. With cobs from seed they seem to be lazy

Click to expand...

cool. I was planning on putting them in a little 2x2.5 side project tent, for mothers or starting seeds or maybe some little breeding projects, something like that.. I have a 5 pack of the ebay tcl strips I was going to throw in there too. they're 154lm/w 3000K and the drivers were included... anyway, I'll get to that eventually, but for now I'm trying to focus on the main area until I get that upgraded and dialed..

Photon Flinger said:

I would suggest you start with a target power budget and build around it. For example, you are running around 2400w plus 350w for qbs if I got that correctly for 49ft2 or roughly 2 5'x5'. Since you are a good grower and plan to increase efficiency, I think a 25w/ft2 target would be desirable.

That puts you at 1250w total.

Click to expand...

I agree. I was planning on 1260w or so total.. for lighting purposes, I think of the 7x7 space as being divided into four 3x3 quadrants by a 1 foot wide + shaped aisle... I was shooting for 35w per sqft. 3 x 3 = 9sqft, 35w/sqft x 9sqft = 315w, 315w x 4 = 1260w... but ya, you could also think of that as about 25w/sqft over the whole 7x7 space.. right now I have 4 qbs at 350w in one of the quadrants and the plants under them are doing great, but I could easily have them dimmed down and closer to the plants... I have them up higher so they aren't at a drastically different height than the blurples, which of course need to be 18-24in minimum from the plants because they are such high wattage...

Photon Flinger said:

Play the balancing game to figure out the most cost effective way you are comfortable with for drivers. Whatever money you have left is going into led and heat sinks. Once again, most cost effective way that works with the drivers you are comfortable with.

When you get more money buy more of whatever light source you chose and put them in parallel on the existing drivers for an efficiency bump resulting in a reduction of heat.

Click to expand...

Ya I think that's a good idea. I'm leaning toward making 4 more lights like the ones I already made to start with.. each has 2 qb288s on a triple heatsink run by an HLG-320H-54A... I can run them close to max at 315w each until I can add 4 more triple heatsinks with 2 qb288s each, then I can double up and run 4 boards per driver in each quadrant instead of just 2...

I need to add some separate veg space too, that's why I'm going to move the 3 qb lights I already built to either 3 2x4 tents or 1 4x4 tent.. they're on the slate 3 triple heatsinks so they dissipate heat better dimmed down, plus of course they're more efficient that way... if I put them in 3 2x4 tents that would give me 24 sqft of extra space.. if I put them in a 4x4 and drive all 3 with only 2 of the drivers, that will give me only 16 sqft of veg space but with more uniform coverage and greater efficiency, plus I can then use the extra driver for one of the main area's lights, so one less driver I have to buy.... still weighing that... my space is short so I have to use short tents... I can get 3 2x4x5 vivosun tents for about $240 total, whereas a 4x4 gorilla shorty (the only short 4x4 tent I know of) is $345... I do like the idea of running 6 boards on 2 HLG-320s in a 4x4 though....

If you are comfortable with the ones you made then it be best to stick with them.

I don't think you can go wrong with any of the options listed just like Cobkits said. Cobs, qbs, they are all white phosphor light sources as Cob puts it, and they roughly perform the same.

You can also look at it in the perspective of total led LES (light emitting surface) possible.

ganjamystic said:

right. they're hardly aisles... by the time the plants fill in I can barely get through, but I can just barely if need be and that's the point. you are correct. all stray light just lights the lowers of the adjacent quadrants...

I would drive all options just hard enough to achieve 315W in each 3x3..

• 2 x 1000w de hps (on light movers, dimmed to 60%)
• 4 x 315w cmh (100%)
  
• 8 x 260w qb288 kits (2 per 3x3, dimmed to ~157.5W each)
• 4 x 315w qb288 diy lights (100%)
• 4 x 315w qb120 diy lights (6 qb120s each, dimmed to ~52.5W each)
• 4 x 315w photonfantom diy lights (4 photon fantom boards at ~79W each)
• 4 x 400w timber 4vs (dimmed to ~315W each)

I grew with hps for 7 years before going led, so I'm familiar with the spectrum.. it works!

from my (admittedly limited) understanding de hps is significantly more efficient than cmh... any reason in particular you'd choose the cmh?.. better spectrum in your opinion? you don't like light movers? the light movers would bring down the efficiency?... care to elaborate a little more on this?

thank you for all your help. I realize this stuff probably seems easy to you by now (since it's your line of work and all), but for me and I'm sure a lot of other growers on here, this stuff is literally like a foreign language... and in my case, not like spanish which I took in highschool and sort of remember, more like chinese or japanese or something that doesn't even use the same letters...

I'm really trying my best to figure shit out myself, but I'd prefer help from the much more knowledgeable than myself peeps on here if anyone's willing..

Click to expand...

hehe, I feel exactly the same; im lost in a sea of options

jlichi said:

hehe, I feel exactly the same; im lost in a sea of options

Click to expand...

Hobby or business?

I think that this is an important question people should be asking when looking at the options.

If you are doing this as a hobby, you really can't go wrong trying different things and have the freedom to change mid grow. A lot less pressure and a whole lot more fun.

Business people want repeatable results and don't often have the freedom to change. A ton of pressure and not so much fun.

Why are none of the board threads talking about chip level stats? If one board has 12 samsungs in parallel and another has 18... and you run them at the same current, which is more efficient....

Growmau5 said:

Why are none of the board threads talking about chip level stats? If one board has 12 samsungs in parallel and another has 18... and you run them at the same current, which is more efficient....

Click to expand...

What are chip level stats?

Im gonna make a noobie attempt at guessing the answer, 12 is more efficient

Unagi said:

12 is more efficient

Click to expand...

nope, it is the board that has 18 parallel strings because the current is divided by 18 so each chip is driven at a lower amperage and therefore it is more efficient than the other board.

tomate said:

nope, it is the board that has 18 parallel strings because the current is divided by 18 so each chip is driven at a lower amperage and therefore it is more efficient than the other board.

Click to expand...

so is the qb304 the most efficient quantum board then?

When you guys are shopping around for boards from who ever. Keep this in mind:
The Samsung lm561c led is a
2.8 volt chip (nominal)
200ma max current

http://www.samsung.com/global/business/business-images/led/file/product/products/201607/Data_Sheet_LM561C_Rev.7.0.pdf

If the board you are interested in is a 36v board, it has 12 or 13 strings of leds in series.
If it's a 54v board , it has around 19 strings in series. These "strings" are parallel traces on the board that divide the current from your driver amongst the leds.
If a board maker recommended a 2.1amp drive current, he must have a minimum of 11 chips in parallel to distribute that 2100ma and not blow these 200ma chips. If a board has 18 chips in parallel per string, with a 2100ma driver, that's 116ma per chip.

In my opinion, this is how people should be comparing a $50 board to a $75 board that both use the same leds.

Growmau5 said:

Why are none of the board threads talking about chip level stats? If one board has 12 samsungs in parallel and another has 18... and you run them at the same current, which is more efficient....

Click to expand...

@Growmau5 , thanks for dropping in man. I wish I understood what you're asking here... I feel like I'm learning a new language.. still in 101 I guess... I'm just trying to figure out the most efficient lighting upgrade for my space without breaking the bank...

Growmau5 said:

When you guys are shopping around for boards from who ever. Keep this in mind:
The Samsung lm561c led is a
2.8 volt chip (nominal)
200ma max current

http://www.samsung.com/global/business/business-images/led/file/product/products/201607/Data_Sheet_LM561C_Rev.7.0.pdf

If the board you are interested in is a 36v board, it has 12 or 13 strings of leds in series.
If it's a 54v board , it has around 19 strings in series. These "strings" are parallel traces on the board that divide the current from your driver amongst the leds.
If a board maker recommended a 2.1amp drive current, he must have a minimum of 11 chips in parallel to distribute that 2100ma and not blow these 200ma chips. If a board has 18 chips in parallel per string, with a 2100ma driver, that's 116ma per chip.

In my opinion, this is how people should be comparing a $50 board to a $75 board that both use the same leds.

Click to expand...

word.. thanks for this info. I think I get what you're saying now

Growmau5 said:

In my opinion, this is how people should be comparing a $50 board to a $75 board that both use the same leds.

Click to expand...

well fundamentally isnt the gross number of diodes more important than how they are laid out?

at the end of the day you have a board... with a bunch of chips.... that are all going to see equal load

if board A has 350 chips and board B has 420 chips (and the boards are of similar size), watt for watt the 420 chip board will always have all of its chips operating at a lower load than the 350 chip board.

the actual current put thru the board (and thus the chips) is totally determined by the user and can span the whole range of efficacy... but watt for watt the 420 chip board should always outperform the 350 regardless of how the dies are laid out

ganjamystic said:

@Growmau5 , thanks for dropping in man. I wish I understood what you're asking here... I feel like I'm learning a new language.. still in 101 I guess... I'm just trying to figure out the most efficient lighting upgrade for my space without breaking the bank...

Click to expand...

I guess what I'm saying is; there is a lot more to pcb design than : how many diodes, and how many watts. Efficiency has always come from the same places.
-drive current
-driver efficiency
-thermal management

Some of the newer start up board makers are claiming high wattage boards, no heatsink needed at a very attractive price. All good things right? If you look close enough, that $50 board may be running these chips at 75-80% of their max current. While the $60 or $75 board is only pushing them at 60% of the leds max current. Sure a consumer can save a couple bucks up front, but they are getting a vastly inferior product, potentially

CobKits said:

well fundamentally isnt the gross number of diodes more important than how they are laid out?

at the end of the day you have a board... with a bunch of chips.... that are all going to see equal load

if board A has 350 chips and board B has 420 chips (and the boards are of similar size), watt for watt the 420 chip board will always have all of its chips operating at a lower load than the 350 chip board.

the actual current put thru the board (and thus the chips) is totally determined by the user and can span the whole range of efficacy... but watt for watt the 420 chip board should always outperform the 350 regardless of how the dies are laid out

Click to expand...

Increasing Gross number of chips can result in two possible outcomes:
A) a higher wattage board
B) a more efficient board.

If you fix the wattage of two 50v boards to say 100w, then obviously the board with more leds will be more efficient. But unfortunately, the marketplace doesn't have fixed variables. And that's my point, people need to learn how to look at the traces, count the strings and estimate the drive current on a per chip basis. That's the only way that I can make an apples to apples comparison between all these Samsung boards.

right but isnt it simpler to say

250 chip board
100W input power

100/250= 0.25W/chip

0.25W / 2.8V = 89 mA/chip

regardless of how the chips are arranged

(obv 2.8V is a simplification but youd still root out the relative info you need)

Growmau5 said:

I guess what I'm saying is; there is a lot more to pcb design than : how many diodes, and how many watts. Efficiency has always come from the same places.
-drive current
-driver efficiency
-thermal management

Click to expand...

all of which are up to the user

Growmau5 said:

Some of the newer start up board makers are claiming high wattage boards, no heatsink needed at a very attractive price. All good things right? If you look close enough, that $50 board may be running these chips at 75-80% of their max current. While the $60 or $75 board is only pushing them at 60% of the leds max current. Sure a consumer can save a couple bucks up front, but they are getting a vastly inferior product, potentially

Click to expand...

same with cobs i suppose. nobody is running clu058s at their nominal wattage of several hundred watts

your analogy is more like somebody who buys an (actual) 200W light, one has 4 cobs, one has 9 of the same cobs. caveat emptor on what the difference is

i feel like were saying the same thing in different ways