Would 1212's measure hotter in the center of the chip than the outer?
- Thread starter Joe34
- Start date
Danielson999
Well-Known Member
Since nobody has given you anything I'll bug ya a bit more.
When we mount cobs on a larger (than the cob) heatsink, the laws of natural convection dictate that the heatsink will cool faster on the outside edges. So since our heatsinks are always hotter the closer you get to the center, one would think that the cob has no option than to be hottest in the center also. The difference in temps would be so minimal you might not be able to measure it.
This is assuming you mount the cob horizontally. If you mounted it vertically the hot spot would shift towards the upper side of the cob because of the heat exchange occurring as the hot air rises over the sink. I think your diagram is spot on.
Why are you interested in knowing specifics about this? just curious...
When we mount cobs on a larger (than the cob) heatsink, the laws of natural convection dictate that the heatsink will cool faster on the outside edges. So since our heatsinks are always hotter the closer you get to the center, one would think that the cob has no option than to be hottest in the center also. The difference in temps would be so minimal you might not be able to measure it.
This is assuming you mount the cob horizontally. If you mounted it vertically the hot spot would shift towards the upper side of the cob because of the heat exchange occurring as the hot air rises over the sink. I think your diagram is spot on.
Why are you interested in knowing specifics about this? just curious...
Last edited:
CobKits
Well-Known Member
id say no. the shortest heat dissipation path from the LES to the heatsink is a line perpendicular to the chip/TIM/heatsink interface. provided that the heatsink is of adequate thickness, this path is essentially the samr fo rall dies. ive youve ever looked closely at a chip, the dies are pretty evenly spread over the LES
Danielson999
Well-Known Member
You can't sidestep thermodynamics. As long as it's mounted horizontally to a heatsink larger than itself, the center is always going to be hotter. Pretend you ran the cob suspended in mid air with no heatsink, convection would suck air from below the cob around the outside edges creating a cooling effect. If you add a heatsink, it's just an extension of the cob and the same principles will apply. Obviously adding forced convection can alter results but the center of your cob is always going to be hottest but I doubt we could measure the difference from one side of the LES to the other. From one side of the heatsink to the other we can definitely measure though.
Danielson999
Well-Known Member
If anybody wants to read the mother of all heatsink books that includes everything from windtunnels, heatpipes, heatsinks, liquid cooling, design, thermal management and everything you could possibly think about....check out Advanced Thermal Solutions Inc book Qpedia Thermal Management – Electronics Cooling Book. It would take a year or two to read it all and an engineering degree might help but it's full of specific info to LED users. Best book out there. There are 4 chapters, each is expensive AF! Might want to check out the content of each because not all of them are necessary for most of us.
Sorry 5 chapters, seen HERE
Sorry 5 chapters, seen HERE
Danielson999
Well-Known Member
Yeah, maybe with a larger cob it would be more evident.
Danielson999
Well-Known Member
Actually it's not the same temp, it's the same color. It's hard to read but the colors actually represent a heat 'range'. I know what you mean though but you still can't stop the heatsink from sucking the heat out toward the edges, leaving the center warmer. Obviously it will be so minimal which is why I originally asked why he wanted to know.
sixstring2112
Well-Known Member
who cares lmao.but seriously,danielson that is some heavy stuff you linked and im still having fun watching that moving pic of the heatsink 
frica
Well-Known Member
I don't think the center of the COB would be the hottest since the diodes inside the board are all spread around evenly. (mostly)
So thermal draw should be spread evenly around the LES.
Or just about.
But if you're talking about the heatsink, the part directly above the LES will be the hottest.
So thermal draw should be spread evenly around the LES.
Or just about.
But if you're talking about the heatsink, the part directly above the LES will be the hottest.
Danielson999
Well-Known Member
Pretty much exactly what we thought. Maybe less than ½ of a ° variance across the LES. But this does show in an excellent way how natural convection cools the outer edges. You can also see the different colors of white across the LES are kind of random, probably just something to do with the thermal paste thickness.