DIY with Chilled PCB's and cxb3590

spcFLETCH

spcFLETCH

Well-Known Member
Airwalker16 said:
@Growmau5 plans on buying one of Vitalys 650W units? Is that 12 (4x3 horz/vert ducted, water cooled) of them? At 50w ec, isn't that 13?
Click to expand...
Have to ask him if he's really going to pick one up. I hope he does. These lights need an expert reviewing them. I'm doing a comparison w/3 Lights.
Btw, the white and reds are rated @1000ma capacity I believe. Separate channels/drivers. So each pcbs is 47.5w @700ma...but since 2 channels can be driven @ 1000ma...you get the jist.
 
Airwalker16

Airwalker16

Well-Known Member
spcFLETCH said:
Have to ask him if he's really going to pick one up. I hope he does. These lights need an expert reviewing them. I'm doing a comparison w/3 Lights.
Btw, the white and reds are rated @1000ma capacity I believe. Separate channels/drivers. So each pcbs is 47.5w @700ma...but since 2 channels can be driven @ 1000ma...you get the jist.
Click to expand...
Ah
It's just with these kits it's all at 700 on the 200W drivers. I wouldn't be too erked running all at 700. More efficient anywho.
 
Humble grower

Humble grower

Well-Known Member
Hey bud thanks for sharing your work I'm definitely interested in seeing how these preform vs cobs. Not sure if you would know the answer to this but do you know how long they will be on sale for? I'd definitely like to get a set of them but just got done dropping 2k on cobs in the last few weeks so I need to get my funds back up a bit
 
spcFLETCH

spcFLETCH

Well-Known Member
Humble grower said:
Hey bud thanks for sharing your work I'm definitely interested in seeing how these preform vs cobs. Not sure if you would know the answer to this but do you know how long they will be on sale for? I'd definitely like to get a set of them but just got done dropping 2k on cobs in the last few weeks so I need to get my funds back up a bit
Click to expand...
Until they are completely gone. =)
 
spcFLETCH

spcFLETCH

Well-Known Member
V, Breaking down the light loss because of Fluence science passive design.

In ALL fluence specifications it has this statement:




Suitable for operation in still air ambient temperatures from -40°C (-40°F) to 35°C (95°F) when suspended and up to 30°C (86°F) when surface mounted. Operating temperature ranges between 25°C (77°F) to 30°C (86°F) above ambient.



So, let' do the math:


1 - Average grow room temps recommended for led lit gardens is 85°F. Let's use the most common mounting method and have the lights suspended from the ceiling (the better case scenario from their statement above vs surface mounted, which gives you an additional 10°F overhead before you reach their max allowed ambient air temp of 95°F).

2 - That puts the lights average operating temperature at 81.5°F above your room's ambient temperature of 85°F, which means their case/enclosure temperature is 166.5°F/ 75°C. (I just added 85°F ambient + 81.5°F average temp rise from their data above).



3 - Now, this is the enclosure/case operating temperature, mind you. Figuring out the LED junction temperature is more tricky since we don't know their lights': heatsink, mcpcb and thermal interface material thermal conductivity - which when combined would allow you to calculate the LED junction temperature. Lets give them a a nice and fair estimate of another 10°C rise from heatsink to junction for a final result of: 85°C junction temperature.




4 - The graph attached is from the datasheet for their white LED they use, the red dot indicates the temperature of the junction their lights would be at under the conditions laid out above. As you can see they are loosing 10% light output/efficiency JUST by the passive cooling's extra heat rise. This, of course, assumes they are running at the manufacturer's test current of 65mA but by judging by the number of chips and stated wattages of their lights they are likely running at least at 100ma per diode, which would skew the flux falloff in that graph by quite a bit, lets assume a loss of 15-20% light loss just on all of the info so far presented.




SO, I guess this is my attempt to show you the "cost" of passive cooling, at least from the perspective of these specific lights, chips and their published specs.
 
P

PurpleBuz

Well-Known Member
spcFLETCH said:
V, Breaking down the light loss because of Fluence science passive design.

In ALL fluence specifications it has this statement:




Suitable for operation in still air ambient temperatures from -40°C (-40°F) to 35°C (95°F) when suspended and up to 30°C (86°F) when surface mounted. Operating temperature ranges between 25°C (77°F) to 30°C (86°F) above ambient.



So, let' do the math:


1 - Average grow room temps recommended for led lit gardens is 85°F. Let's use the most common mounting method and have the lights suspended from the ceiling (the better case scenario from their statement above vs surface mounted, which gives you an additional 10°F overhead before you reach their max allowed ambient air temp of 95°F).

2 - That puts the lights average operating temperature at 81.5°F above your room's ambient temperature of 85°F, which means their case/enclosure temperature is 166.5°F/ 75°C. (I just added 85°F ambient + 81.5°F average temp rise from their data above).



3 - Now, this is the enclosure/case operating temperature, mind you. Figuring out the LED junction temperature is more tricky since we don't know their lights': heatsink, mcpcb and thermal interface material thermal conductivity - which when combined would allow you to calculate the LED junction temperature. Lets give them a a nice and fair estimate of another 10°C rise from heatsink to junction for a final result of: 85°C junction temperature.




4 - The graph attached is from the datasheet for their white LED they use, the red dot indicates the temperature of the junction their lights would be at under the conditions laid out above. As you can see they are loosing 10% light output/efficiency JUST by the passive cooling's extra heat rise. This, of course, assumes they are running at the manufacturer's test current of 65mA but by judging by the number of chips and stated wattages of their lights they are likely running at least at 100ma per diode, which would skew the flux falloff in that graph by quite a bit, lets assume a loss of 15-20% light loss just on all of the info so far presented.




SO, I guess this is my attempt to show you the "cost" of passive cooling, at least from the perspective of these specific lights, chips and their published specs.
Click to expand...
very cool analysis. Problem is your assuming how many degrees over ambient they are targeting. Need to know the actual Tc or a projection based off a temp measurement. Could you stick a probe in it and take its temperature ?
 
PSUAGRO.

PSUAGRO.

Well-Known Member
Bad thermal design/management by fluence..........doesn't mean passive cooling is bad

Osram zelion was never hot or even warm with 85+f room temps, they ran the ssl 80 chips underspec(340ma), beefy mcpcb, big anodized heat sink , proper diode spacing. Drivers placed on the perimeter of the aluminum shell======= good engineering, das German way!lol


Edit, Osram ssl 80 also have some of the best jt/attenuation #s in the industry , so that helps too
 
Last edited:
spcFLETCH

spcFLETCH

Well-Known Member
PSUAGRO. said:
Bad thermal design/management by fluence..........doesn't mean passive cooling is bad

Osram zelion was never hot or even warm with 85+f room temps, they ran the ssl 80 chips underspec(340ma), beefy mcpcb, big anodized heat sink , proper diode spacing. Drivers placed on the perimeter of the aluminum shell======= good engineering, das German way!lol
Click to expand...
Just bad design by this company. Not bad for passive cooling. Getting some pin sinks from Pacific Light concepts in a couple weeks=)
 
spcFLETCH

spcFLETCH

Well-Known Member
PurpleBuz said:
very cool analysis. Problem is your assuming how many degrees over ambient they are targeting. Need to know the actual Tc or a projection based off a temp measurement. Could you stick a probe in it and take its temperature ?
Click to expand...
I will keep it in mind once the experiment is going. The #s were based off my plan to run these lights in a controlled environment in 85F. I will try and remember. They will be in a tent in about 3 weeks.
 
spcFLETCH

spcFLETCH

Well-Known Member
spcFLETCH said:
I will keep it in mind once the experiment is going. The #s were based off my plan to run these lights in a controlled environment in 85F. I will try and remember. They will be in a tent in about 3 weeks.
Click to expand...
By the way, That was not my work just posting his breakdown.
 
spcFLETCH

spcFLETCH

Well-Known Member
PurpleBuz said:
very cool analysis. Problem is your assuming how many degrees over ambient they are targeting. Need to know the actual Tc or a projection based off a temp measurement. Could you stick a probe in it and take its temperature ?
Click to expand...
What do you think of the concept? Can we say that the theory is probably correct? Once we measure the temps, do you think there is reason to believe that 15-20% loss is viable? Therefore 2.2 per joule is more like 1.76 (20% loss)
 
P

PurpleBuz

Well-Known Member
spcFLETCH said:
What do you think of the concept? Can we say that the theory is probably correct? Once we measure the temps, do you think there is reason to believe that 15-20% loss is viable? Therefore 2.2 per joule is more like 1.76 (20% loss)
Click to expand...
probably ... while they publish 2.2 per joule, it seems really hard to achieve that with the physical size for passive cooling. We will know once you take its temperature.

btw thanks for the show and tell!
 
You must log in or register to reply here.
Top