Limo's CXB3590 build

SupraSPL

SupraSPL

Well-Known Member
Growmau5 said:
I was doing the same thing for a while. since in supras chart at the top of each profile it says : "in or cm - in" every calculation I did, i was like " christ, thats a f*cking huge heatsink" lol
Click to expand...
@Abiqua @alesh Sorry for the confusion there, the "in or cm" box is allows me to enter the heatsink dimensions using in or cm. The spreadsheet is interactive and maybe does not translate well as a reference sheet. It is designed so I can quickly change the heatsink length (normally in inches) to achieve the desired surface area. As a reference sheet it only shows the results for 1" of length, the idea being that users can multiply as needed.

I prefer to think in terms of cm² for surface area, but since heatsinkUSA is known by inches of length, that is how we end up with the awkward conversion. It is just as annoying to me as it must be to metric system users, I am certain our system of measurement makes angels cry. I will post 2 reference sheets, one entirely in inches and one in cm to end that confusion
 
Last edited:
SupraSPL

SupraSPL

Well-Known Member
guod said:
BTW.
placing cobs near the Edges and Corner of a heatsink also shows me that the Supra Lednurse has no Idea how cooling with Heatsinks really works!
https://www.rollitup.org/t/heatsinks-for-diy-led-lamps.851907/page-55#post-11896376
Click to expand...
Guod, I make plenty of mistakes and I genuinely appreciate when people catch them and straighten me out, but you just seem pissed off about something? Anyway, the reason the COBs are staggered in this example is to improve uniformity in the canopy. True, it is less than perfect heat load balancing but the difference in output will be negligible, .25%? Don't believe me, test it for yourself. The improvement in uniformity is more significant, it is a compromise. If you want perfect cooling and perfect uniformity you would have to use CPU coolers and this DIYer did not want that setup. So if you have a better solution for him please enlighten us (on the relevant thread not this one)

guod said:
The published extrusion data shows natural convection performance for a three inch section with a centrally located point source heat load. etc etc
Click to expand...
For the most part I ignore C/W/3" data, even with length correction and temp corrections factors. The standard "natural convection" data is based on a 75C difference between the heatsink temp and ambient AND it is vertically oriented. That probbaly creates some decent air convection. But that metric is based on an application that has little resemblance to ours. We are trying to keep heatsink temps much closer to ambient, normally horizontally oriented and benefiting from a slight circulation fan breeze. That is why we use surface area as a guide instead, KNNAs approach from way back.

In my mind the thing that will help us the most is temp droop testing on the actual heatsinks, which I have done plenty of. I adjusted his recommendations to 40cm²/ heat W for active and 120cm²/ heat W for passive, subject to change and open to opinions.
 
Last edited:
SupraSPL

SupraSPL

Well-Known Member
Limosnero said:
40" of the 5.375 profile...gonna do a 3-2-3 on 3 heat sinks...eventually add another heatsink & more COBs for a 4-4-4-4 configuration.
Click to expand...
Limosnero said:
Hmmm....could someone please layout the correct formula? I was calculating that for passive cooling.
234.78 x 91.44 = 21468.2832/110 = 195.17
Click to expand...
Nice job on the build man, I think it will treat your ladies right :)

So 40" of the 5.375" = 13829cm², which is good for 115 Watts of heat on a passive cooling friendly heatsink profile. The 5.375" is not the most passive cooling friendly due to close fin spacing, so we might want to use a bit more cm² than the normal recommendation of 120cm²/W of heat.

So if you installed 4 CXB3590s (196W) that is 86W of heat into the heatsink. 13829cm²/86W = 161cm²/W of heat. That is quite a bit of surface area and I think that will give you very good temp droop figures so no worries when/if you upgrade to 4 COBs/ heatsink. That said, you could save some weight and some $ by using 5.88"X40" if you decide to order the fourth heatsink.
 
Limosnero

Limosnero

Well-Known Member
SupraSPL said:
@Limosnero

CXB3590 3500K @ 1.4A = ~49W and 56.3% efficiency
So 8 of them = 392 dissipation W * .563 = 221 PAR W
221 * .85 wall losses = 188 PAR W
188/16ft² = 11.7 PAR W/ft²
11.7 * 4.65umol/s/W * 10.7 = 583 PPFD averaged

Very much looking forward to see how the ladies like this tent :leaf:
Click to expand...
Thank you for running those #s & for stopping in SupraSPL...it is appreciated.

I did wind up with 3 - 36" lengths of the 5.375" profile. When taking heat measurements, I'm getting about 60 degrees Celsius...& really that's just blindly shooting the laser for the hottest spot as the Tc is covered by the ideal holder. However I will still add a small oscillating clip fan at the top of the tent to point directly at the heat sinks. The spec sheet shows 85c @ 3600mA as the maximum current rating. Although it states something about assumed thermal management.
 
littlejacob

littlejacob

Well-Known Member
Bonjour
Lmfao!!!
You gave me a good idea...I am hairdresser so I am going to try to cut my customers hair this way...it look easier and faster than with my cisor and razor...lol
Anyway it is a good idea...I love it!
CU
 
You must log in or register to reply here.
Top