advanced theories and questions

banke1

Well-Known Member
clarify some things for me please.

First things first, I'd like to thank you all for the knowledge that is being passed around. I've lurked on these forums for years, and LEDs finally got me to ask some questions.

My goal is to increase electrical efficiency WITHOUT sacrificing flower quality, perhaps even improving it. If I can grow the same top-shelf quality meds with less watts, I'm a happy man.

Try to read everything before you comment because this will come around full circle in the end, sorry for being long winded.

Now, I've read somewhere that in order to maximize efficiency, it would be better to run individual diodes. (is this driver efficiency or umol/J efficiency we are speaking about here?) I remember on this forum somewhere someone said (lol) that you would need an engineering degree to do so, because there are no drivers available for this type of application, and you would have to create your own drivers. Keeping this in mind, I'd also speculate that running individual diodes would give you better canopy coverage (even though 1 cob per sq foot is better than a single hps, one would think that 136 individual LEDS per sq ft would be the cats meow (this is how many LEDS the spydrx+ has...2176 total over a 4x4)
So, better effiency AND more uniform coverage? Hold that thought...


i was planning on going with 1 cob / sq ft, passively cooled with those nice anodized black heat sinks for a dedicated flowering room.

(4) CXB3590 32v CD 3500K @ 1050ma per (1) MW HLG-185H-C1050B

Now as I understand, I would want to be able to dim it, which is why I have selected the B version driver. I am confused on how much dimming is possible, and how exactly this happens.
I know there are 2 wires that come off the driver, but what do they attach to, a knob?
Can I dim the 4 cobs on that 1 driver all down to 700ma? down to 800ma like captain? down to 967.281445740000012ma? down to 350ma? What is the range of dimming available?
I see people dim their 700ma down to 350ma, I'm assuming thats similar, but theres no info on how necessarily.
When I spoke with Fluence about using their SoLunar controller to dim the SpydrX+, they said it is a linear dim that (as far as he knows) doesnt alter the fV so the efficiency of the spydrx+ stays the same at 10% power or 100% power. So this is a different dimming method than running 1050ma cobs @ 900ma?
And if I have 4 bars over a 4x4, do I need 4 separate dimmers with 4 separate knobs, or is there a way to make sure everything is outputting exactly the same amount of light?


Correct me if I am wrong but if I did four of those bars over a 4x4 area, it would cost me ~
$1,300 in cobs/holders/adapters/reflectors (if i got 16 of the mau5 kits from cutter)
$ 325 in heatsinks (assuming i just went with 5.886x48" extruded aluminum from hsusa)
$ 255 in drivers (from mouser)
that totals $1,880
then theres misc expenses of wires, plugs, dimmer(s)?, screws, etc, so I could easily see this light end up being $2,000 to create.
not to mention I would want to purchase things like the drill press, and whatever else I would need to make this build look and BE professional. But lets just call it $2,000 per 4x4.


So, please explain to me why I wouldn't just buy the new SPYDRx? Its apparantly cheaper with better efficiency and coverage, with the added convenience of only 6" distance from canopy, lightweight, and clean wiring.
Am i missing something? I was very excited to build my own lights but when I look at it this way, I feel like they've got me beat, despite the fact that they have a warranty and create this thing with robotics, as opposed to me with 0 knowledge only what I've learned on forums and youtube.

The cobs I was planning on running (again thats CXB3590 CD 3500K 36v) would be at 35.5w each @ 1050ma, correct? so a total of 568w or 35.5w/sqft
the spydrx+ has 685w total / 16sqft = 42.8w/sqft.
So I could run the spydrx+ @83% and it would be drawing the same from the wall as my homemade unit @100%?
But then doesn't that mean the spydrx+ is @ ~754ppfd (.83% of 909) drawing 568w VS the homemade unit @100% and 1050ppfd drawing 568w?
this doesnt coincide with my first statement about how individual diodes are more efficient then vs cobs. im so confused...

What about comparing it to the current top end Veg-Cob-setup using the cxb3590 5000K DB bins @ 350ma? Lol... I digress...
=============================================================

PS. I just got done chewing Aaron from Fluence's ear off.
heres a few things that came out of the convo...
-the physiospec spectrum is replacing ALL of their old BML spectrums and it is ~3500K and 85cri
-they are releasing individual bars based off the spydrx+ bars... no word on price/date as of yet. but they will be available in a few different lengths and dimmable using the physiospec
-theres currently ~5 week build/delivery time expected as of right now on the spydrx+ (didnt ask about the vyprx, sorry)
-the SoLunar controller that you can find on the old BML website can dim up to 3 of the new spydrx+, this is the PWM controller they mention on the fluence website without any link for it
-their heatsinks are anodized
-you can take one or multiple bars out of the spydrx and the rest of it will all still work
-he wasnt 100% but he thinks the spydrx+ uses two drivers.

==========================================================

For reference can someone please fill in the *blanks here and make sure im correct with what i have so far?

with 36v CD 3500K cxb3590 1 cob/ sq ft:

______ _______Watts/cobb _________ ~PPFD_________umol/J

350ma _________ * __________ * __________ *
700ma _________23? _________ * _________*
1050ma _________35.5? ______1050?_________ *
1400ma _________50? _________ * _________ *

with 36v DB 5000K cxb3590 1cob/sqft

_____ _________Watts/cobb_________~PPF D_________ umol/J

350ma _________ * _________ * _________ *
700ma _________ 23? _________ * _________ *
1050ma _________ 35.5? _________ * _________ *
1400ma _________ 50? _________ * _________ *

=======================================

And for over a 128 site clone machine, whats the current(2016) top-tier? (2) 5000k cxb3590 DB 36v @ 350ma sound good? I think this will ultimately be my first actual purchase of LEDs to replace my T5s.

=================================
So in conclusion look, I appreciate all the knowledge brought forth by this community esp the ones who put in all this work to get to this point (growmau5, supra, knna, dgc, thc, cbd, realstyles, greengenes, lucas, spurr, danko, subcool, vader, emery, herer, mom, dad, and my pet turtle and anyone else I've forgot)

But I feel as if theres just got to be someone who really does it "official tissue" that I wouldn't and shouldn't be able to compete with on a DIY level. And hey, I probably wouldn't even be looking into lights by companies like fluence if it werent for you guys shedding light on what is possible.

So, for my veg, mom, and clone lights, I believe I will definitely be creating my own (only because their physiospec individual bars arent out yet, lol), and I want to reduce my power consumption immediately.
But when it comes down to the flowering, I think I'd rather put my trust in a company with a product like that, unless you guru's are able to convince me otherwise.

==========

OH YEAH! and if the majority now believe that full spectrum lights are where its at for plants, and green light is included in that; then is using green lights during the night cycle still OK? LOL. (imagines a world where growers are now using night vision goggles)

==========

thoughts.... from my brain... enter the matrix... takes dab... can relax now...
 
Last edited:

AquariusPanta

Well-Known Member
clarify some things for me please.

First things first, I'd like to thank you all for the knowledge that is being passed around. I've lurked on these forums for years, and LEDs finally got me to ask some questions.

My goal is to increase electrical efficiency WITHOUT sacrificing flower quality, perhaps even improving it. If I can grow the same top-shelf quality meds with less watts, I'm a happy man.

Try to read everything before you comment because this will come around full circle in the end, sorry for being long winded.

Now, I've read somewhere that in order to maximize efficiency, it would be better to run individual diodes. (is this driver efficiency or umol/J efficiency we are speaking about here?) I remember on this forum somewhere someone said (lol) that you would need an engineering degree to do so, because there are no drivers available for this type of application, and you would have to create your own drivers. Keeping this in mind, I'd also speculate that running individual diodes would give you better canopy coverage (even though 1 cob per sq foot is better than a single hps, one would think that 136 individual LEDS per sq ft would be the cats meow (this is how many LEDS the spydrx+ has...2176 total over a 4x4)
So, better effiency AND more uniform coverage? Hold that thought...


i was planning on going with 1 cob / sq ft, passively cooled with those nice anodized black heat sinks for a dedicated flowering room.

(4) CXB3590 32v CD 3500K @ 1050ma per (1) MW HLG-185H-C1050B

Now as I understand, I would want to be able to dim it, which is why I have selected the B version driver. I am confused on how much dimming is possible, and how exactly this happens.
I know there are 2 wires that come off the driver, but what do they attach to, a knob?
Can I dim the 4 cobs on that 1 driver all down to 700ma? down to 800ma like captain? down to 967.281445740000012ma? down to 350ma? What is the range of dimming available?
I see people dim their 700ma down to 350ma, I'm assuming thats similar, but theres no info on how necessarily.
When I spoke with Fluence about using their SoLunar controller to dim the SpydrX+, they said it is a linear dim that (as far as he knows) doesnt alter the fV so the efficiency of the spydrx+ stays the same at 10% power or 100% power. So this is a different dimming method than running 1050ma cobs @ 900ma?
And if I have 4 bars over a 4x4, do I need 4 separate dimmers with 4 separate knobs, or is there a way to make sure everything is outputting exactly the same amount of light?


Correct me if I am wrong but if I did four of those bars over a 4x4 area, it would cost me ~
$1,300 in cobs/holders/adapters/reflectors (if i got 16 of the mau5 kits from cutter)
$ 325 in heatsinks (assuming i just went with 5.886x48" extruded aluminum from hsusa)
$ 255 in drivers (from mouser)
that totals $1,880
then theres misc expenses of wires, plugs, dimmer(s)?, screws, etc, so I could easily see this light end up being $2,000 to create.
not to mention I would want to purchase things like the drill press, and whatever else I would need to make this build look and BE professional. But lets just call it $2,000 per 4x4.


So, please explain to me why I wouldn't just buy the new SPYDRx? Its apparantly cheaper with better efficiency and coverage, with the added convenience of only 6" distance from canopy, lightweight, and clean wiring.
Am i missing something? I was very excited to build my own lights but when I look at it this way, I feel like they've got me beat, despite the fact that they have a warranty and create this thing with robotics, as opposed to me with 0 knowledge only what I've learned on forums and youtube.

The cobs I was planning on running (again thats CXB3590 CD 3500K 36v) would be at 35.5w each @ 1050ma, correct? so a total of 568w or 35.5w/sqft
the spydrx+ has 685w total / 16sqft = 42.8w/sqft.
So I could run the spydrx+ @83% and it would be drawing the same from the wall as my homemade unit @100%?
But then doesn't that mean the spydrx+ is @ ~754ppfd (.83% of 909) drawing 568w VS the homemade unit @100% and 1050ppfd drawing 568w?
this doesnt coincide with my first statement about how individual diodes are more efficient then vs cobs. im so confused...

What about comparing it to the current top end Veg-Cob-setup using the cxb3590 5000K DB bins @ 350ma? Lol... I digress...
=============================================================

PS. I just got done chewing Aaron from Fluence's ear off.
heres a few things that came out of the convo...
-the physiospec spectrum is replacing ALL of their old BML spectrums and it is ~3500K and 85cri
-they are releasing individual bars based off the spydrx+ bars... no word on price/date as of yet. but they will be available in a few different lengths and dimmable using the physiospec
-theres currently ~5 week build/delivery time expected as of right now on the spydrx+ (didnt ask about the vyprx, sorry)
-the SoLunar controller that you can find on the old BML website can dim up to 3 of the new spydrx+, this is the PWM controller they mention on the fluence website without any link for it
-their heatsinks are anodized
-you can take one or multiple bars out of the spydrx and the rest of it will all still work
-he wasnt 100% but he thinks the spydrx+ uses two drivers.

==========================================================

For reference can someone please fill in the *blanks here and make sure im correct with what i have so far?

with 36v CD 3500K cxb3590 1 cob/ sq ft:

______ _______Watts/cobb _________ ~PPFD_________umol/J

350ma _________ * __________ * __________ *
700ma _________23? _________ * _________*
1050ma _________35.5? ______1050?_________ *
1400ma _________50? _________ * _________ *

with 36v DB 5000K cxb3590 1cob/sqft

_____ _________Watts/cobb_________~PPF D_________ umol/J

350ma _________ * _________ * _________ *
700ma _________ 23? _________ * _________ *
1050ma _________ 35.5? _________ * _________ *
1400ma _________ 50? _________ * _________ *

=======================================

And for over a 128 site clone machine, whats the current(2016) top-tier? (2) 5000k cxb3590 DB 36v @ 350ma sound good? I think this will ultimately be my first actual purchase of LEDs to replace my T5s.

=================================
So in conclusion look, I appreciate all the knowledge brought forth by this community esp the ones who put in all this work to get to this point (growmau5, supra, knna, dgc, thc, cbd, realstyles, greengenes, lucas, spurr, danko, subcool, vader, emery, herer, mom, dad, and my pet turtle and anyone else I've forgot)

But I feel as if theres just got to be someone who really does it "official tissue" that I wouldn't and shouldn't be able to compete with on a DIY level. And hey, I probably wouldn't even be looking into lights by companies like fluence if it werent for you guys shedding light on what is possible.

So, for my veg, mom, and clone lights, I believe I will definitely be creating my own (only because their physiospec individual bars arent out yet, lol), and I want to reduce my power consumption immediately.
But when it comes down to the flowering, I think I'd rather put my trust in a company with a product like that, unless you guru's are able to convince me otherwise.

==========

OH YEAH! and if the majority now believe that full spectrum lights are where its at for plants, and green light is included in that; then is using green lights during the night cycle still OK? LOL. (imagines a world where growers are now using night vision goggles)

==========

thoughts.... from my brain... enter the matrix... takes dab... can relax now...
:hug:
 

banke1

Well-Known Member
so my first post got a hug?

i literally only started to learn about led's like 2 weeks ago
 

Rahz

Well-Known Member
If running the emitters at 60% efficiency is important to you then you will end up paying more up front (though perhaps still less than a retail unit). This will either result in higher PPF or same PPF for less electricity, in which case the saving stack up over time in the form of the electrical bill. If the emitters aren't known you would need to inquire about the emitter efficiency (output efficiency) and compare that to your DIY projections. PPFD values derived from spot measurements can't be directly compared to a PPFD measurement derived from a datasheet.

Softly driven cobs can get pretty close to the canopy. At 1.4 amps 8-9 inches is reasonable. At 1.05 amps you will be able to get to around 6-7 inches.

You can run 4 CXBs at 1.05 amps on a HLG-120H-C1050B.

If you found that the retail lamp you're looking at has less than 60% efficiency, you can go back and make a comparison with a higher drive current. Savings would become more apparent.
 

PurpleBuz

Well-Known Member
you can cut your cost down a little by deploying 3 4 foot bars in a 4x4 instead of 4.
1 fully packed four foot bar of cxb3590s will cover a 5x2 area. an hlg-185h-C1050 can handle 5 cxb3590s.
3 bars in a 4x4 gives enough overlap between bars for very even coverage.

Active cooling can be more efficient than passive cooling. One good 140mm fan in the center of a 4' 5.8 heatsinkusa profile runs more efficiently than without a fan, My current favorite is the rosewill hyperborea fan powered by a meanwell APV-12 .

Meanwell has a few different dimming methods.
The A version allows you to dim each bar\driver with an internal pot.
The B version has two wires for an external pot (you supply). The wires from multiple drivers can be controlled from one pot.

if you shop around you should be able to drive your costs down significantly...

the new spydrx looks like an awesome light, the upcoming individual bars look even better. I haven't seen many grows from the new lights yet, but on paper has lots of potential.
 

welight

Well-Known Member
clarify some things for me please.


Correct me if I am wrong but if I did four of those bars over a 4x4 area, it would cost me ~
$1,300 in cobs/holders/adapters/reflectors (if i got 16 of the mau5 kits from cutter)
$ 325 in heatsinks (assuming i just went with 5.886x48" extruded aluminum from hsusa)
$ 255 in drivers (from mouser)
that totals $1,880
then theres misc expenses of wires, plugs, dimmer(s)?, screws, etc, so I could easily see this light end up being $2,000 to create.
not to mention I would want to purchase things like the drill press, and whatever else I would need to make this build look and BE professional. But lets just call it $2,000 per 4x4.


..
MAU5 will come with predrilled and tapped heatsinks to suit the ideal connectors and drivers
Cheers
Mark
 
Top