Thinking of a new light ..
- Thread starter stardustsailor
- Start date
Hgrow
Active Member
Nice Work.Your heat sink is very impressive . How's the temperature of your led light when working ?Yes ,Cobs make things wayyyyyy easier ....
Yeah,I sometimes can't get over it myself,when I look at them !
And they still work fine ( Although I've managed to destroy one of the pcbs ,in one panel,by accidently decapitaning
the lens dome of couple of Oslons on that pcb ...Tiny and very sensitive ...)
Reflowed ,at stove ,twelve Oslons at once per mcpcb ....
9x mcpcbs ....108 Oslons hot-plate reflowed...
Yeap...It was 'challenging ' enough !!!
But ,for sure they're great little ones ,those Oslons !
View attachment 3151007
View attachment 3151008
stardustsailor
Well-Known Member
Output current range set !
Minimum set is @ 350mA .
( ~12 Watts/CXA3070 )
Max settting is @ ~ 2010mA ( 40 V )
( ~80 Watts / CXA3070 )
Minimum set is @ 350mA .
( ~12 Watts/CXA3070 )
Max settting is @ ~ 2010mA ( 40 V )
( ~80 Watts / CXA3070 )
stardustsailor
Well-Known Member
And testing the CXA case over-temperature protection.
Tc Limit set at 45°C for ease at testing .
( Normal operational limit is @55°C . )
Chip #3 ,of course ,was the first to go over the limit ....
Either I 've to check sensors #2 & #4 ...
Or either the chips ( installation / TIM ) #3 & #1 ....
There's an difference of 3-8°C ,between those pairs ...
CXA3070 driven at 2 Amps .
Tc Limit set at 45°C for ease at testing .
( Normal operational limit is @55°C . )
Chip #3 ,of course ,was the first to go over the limit ....
Either I 've to check sensors #2 & #4 ...
Or either the chips ( installation / TIM ) #3 & #1 ....
There's an difference of 3-8°C ,between those pairs ...
CXA3070 driven at 2 Amps .
Last edited:
stardustsailor
Well-Known Member
Time for 'burn-in ' .....
stardustsailor
Well-Known Member
Yes .It definitely feels like plenty of heat,coming from the COBs on the light side...
Besides the chips,actually heatsink is pretty cool!
TC #3 is at ~36.5 °C ,with fan speed at 4th/12 speed setting ( ~ 8V ) and
CXAs driven somewhere between >500mA - < 1000mA....
(That's why an ampmeter is needed ,after all...
...)
Steady for more than 6 hours now ...
Cooling at it's best ,I guess ...
But ,not 'silent' ,even when at 4/12 of total available fan power...
..At least for my 'standards' ....
Those Sunons aren't that silent ...
Last edited:
stardustsailor
Well-Known Member
The theory behind cooling "GameChanger" :
One crucial point of cooling using a HSF (heatsink-fan) cooling system ,is
to manage to direct the coolest part of air onto the hottest part of heatsink .
.....
That is actually one of few basic laws of an efficient HSF cooling system .
Blowing or sucking air ?
There's not a definite answer to that question ...
It depends of the system to be cooled ,which of those two airflow directions ,
should be preferred over the other .
In the design of "GameChanger" sucking air was preferred over blowing air onto the heatsink .
Each heatsink is quite flat with a 10mm thick base .
Thick base on a heatsink means high heat surface dispersion .
As heat 'travels' through base ,cause of thickness it 'disperses' more ,
when it reaches the back side of the heatsink (where the fins are ) ...
Fins are quite spaced apart ,being thick and short .
They 'suck' plenty of heat fast and have the ability to 'transfer' it fast ...
To convect it with air.
But they are not the kind of fins to convect better when air is blown to them.
They are actually "passive" efficient fins .
If air is blown to a back side of a heatsink ,at least two things are happening :
A ) The air comes 'in contact' with the fins first ,thus cooling them .
Kinda 'locally' ...So in this design i) heat generator / chip should be placed exactly beneath airflow.There's where's the highest convection .(cooling ).
As air reaches the bottom of the heatsink ,is rather hot with not much "heat capacitance" to efficiently convect heat as before ....(coming out from fan blades ) ....
So base of the heatsink is not effiently cooled as the fins .
But because fins are cooled first (more efficiently ) ,more heat goes to them from base of the heatsink ...
So in this design ii) Fins made specificly for blowing air to them are high and thin ,oftenly closely spaced between them ,like in car/fridge "radiators" ...
B ) Dust/debris gathers on front , where fan is blowing .
Another type of heatsink ...
Completely another type ...
So at each heatsink ,two CXA3070 are placed near the sides ( length-wise ) of the heatsink .
Hottest point is near the sides ...(length-wise...)
In order ,to keep as much possible,of the incoming (sucked ) cool air ,lo, near the thick base and
short /thick fins ,specially at both the sides of heatsink's length...
A 'restrainer' is used to 'constrain' airflow where is needed most and when is at it's coolest point .
So..A wide fan is used at each heatsink and a restrainer lid ,with the fan centered at back of each heatsink.
Sucking air .Cool air enters at both sides ,length-wise,and at it's coolest state ,convects with the hottest part of the heatsink.Where CXA's are placed.Constrained as is the airflow ,travels to the center of the heatsink (the least convecting part in this design ,as opposed to 'blowing' design ) ,where it exausts in the inner compartment of the case ...
If case did not had a top lid ,then two fans would have been enough ....
But this ,I'm afraid ain't a 'cabriolet' model ...
So another pair of fans is used to exaust the hot air of the inner case compartment...
Because ,it is' in series ' with the heatsink fan pair ,it increases the total ' negative pressure' ...
Sucking air becomes more 'easy' ,but actual airflow has not changed/remains same ...
Just the airflow's 'ability' to 'overcome' 'obstacles',as travels from inlets to exausts ,
is increased/augmented .
Dust will gather firstly(and mostly ) at the air inlets .(all possible ones ...)
This design has plenty of cooling "power " ..
Handling the heat from the 2A max drive from 4x CXA3070,is probably a piece of cake ...
But ..with a price ...
Noise ...
Quite a lot over the 6/12 speed set ...
One crucial point of cooling using a HSF (heatsink-fan) cooling system ,is
to manage to direct the coolest part of air onto the hottest part of heatsink .
.....
That is actually one of few basic laws of an efficient HSF cooling system .
Blowing or sucking air ?
There's not a definite answer to that question ...
It depends of the system to be cooled ,which of those two airflow directions ,
should be preferred over the other .
In the design of "GameChanger" sucking air was preferred over blowing air onto the heatsink .
Each heatsink is quite flat with a 10mm thick base .
Thick base on a heatsink means high heat surface dispersion .
As heat 'travels' through base ,cause of thickness it 'disperses' more ,
when it reaches the back side of the heatsink (where the fins are ) ...
Fins are quite spaced apart ,being thick and short .
They 'suck' plenty of heat fast and have the ability to 'transfer' it fast ...
To convect it with air.
But they are not the kind of fins to convect better when air is blown to them.
They are actually "passive" efficient fins .
If air is blown to a back side of a heatsink ,at least two things are happening :
A ) The air comes 'in contact' with the fins first ,thus cooling them .
Kinda 'locally' ...So in this design i) heat generator / chip should be placed exactly beneath airflow.There's where's the highest convection .(cooling ).
As air reaches the bottom of the heatsink ,is rather hot with not much "heat capacitance" to efficiently convect heat as before ....(coming out from fan blades ) ....
So base of the heatsink is not effiently cooled as the fins .
But because fins are cooled first (more efficiently ) ,more heat goes to them from base of the heatsink ...
So in this design ii) Fins made specificly for blowing air to them are high and thin ,oftenly closely spaced between them ,like in car/fridge "radiators" ...
B ) Dust/debris gathers on front , where fan is blowing .
Another type of heatsink ...
Completely another type ...
So at each heatsink ,two CXA3070 are placed near the sides ( length-wise ) of the heatsink .
Hottest point is near the sides ...(length-wise...)
In order ,to keep as much possible,of the incoming (sucked ) cool air ,lo, near the thick base and
short /thick fins ,specially at both the sides of heatsink's length...
A 'restrainer' is used to 'constrain' airflow where is needed most and when is at it's coolest point .
So..A wide fan is used at each heatsink and a restrainer lid ,with the fan centered at back of each heatsink.
Sucking air .Cool air enters at both sides ,length-wise,and at it's coolest state ,convects with the hottest part of the heatsink.Where CXA's are placed.Constrained as is the airflow ,travels to the center of the heatsink (the least convecting part in this design ,as opposed to 'blowing' design ) ,where it exausts in the inner compartment of the case ...
If case did not had a top lid ,then two fans would have been enough ....
But this ,I'm afraid ain't a 'cabriolet' model ...
So another pair of fans is used to exaust the hot air of the inner case compartment...
Because ,it is' in series ' with the heatsink fan pair ,it increases the total ' negative pressure' ...
Sucking air becomes more 'easy' ,but actual airflow has not changed/remains same ...
Just the airflow's 'ability' to 'overcome' 'obstacles',as travels from inlets to exausts ,
is increased/augmented .
Dust will gather firstly(and mostly ) at the air inlets .(all possible ones ...)
This design has plenty of cooling "power " ..
Handling the heat from the 2A max drive from 4x CXA3070,is probably a piece of cake ...
But ..with a price ...
Noise ...
Quite a lot over the 6/12 speed set ...
Last edited:
stardustsailor
Well-Known Member
And some power measurements ...
Lowest power at plug : Initial 82 Watt .
Highest power at plug: Initial 410 Watt ,stabilises around 402 Watt .
Chip #3 has for sure to be checked .'
At max setting ,it has a ΔΤ with the rest of the chips of about 10°C .
Something's wrong with that one ...
Lowest power at plug : Initial 82 Watt .
Highest power at plug: Initial 410 Watt ,stabilises around 402 Watt .
Chip #3 has for sure to be checked .'
At max setting ,it has a ΔΤ with the rest of the chips of about 10°C .
Something's wrong with that one ...
PSUAGRO.
Well-Known Member
Interesting info on your cooling methods SDS, remember though that in refrigerators/cars(when not in motion) the air is PULLED through the fins because it also cools the compressor/motor behind the fan at the same time=== cheaper design, best cooling method??IDK,
yeah that a significant temp diff on #3 cob^^ wonder if the "HOLDER" is the issue?? not everyone running cobs has the luxury of checking temps real time like you, well a cheap IR temp gun would be accurate enough to check for a flaw I guess.
Your panel is just damn sexy, can't wait for the grow under it..............be safe grower
yeah that a significant temp diff on #3 cob^^ wonder if the "HOLDER" is the issue?? not everyone running cobs has the luxury of checking temps real time like you, well a cheap IR temp gun would be accurate enough to check for a flaw I guess.
Your panel is just damn sexy, can't wait for the grow under it..............be safe grower
stardustsailor
Well-Known Member
Yes ..
Car radiator fins are horizontally placed ...
And airflow can run throught them ,both ways ...
Pretty useful (for a ..moving... car's radiator ) and equally efficient ,both ways ...
stardustsailor
Well-Known Member
Pet,please 'break your question in pieces ' and repeat ..
I'm sorry,what ?
froghunter
Active Member
congratz my brother. really inspiring job you did.
i hope you are fine in other aspects of your life as well
(i lurk silently like a ninja
)
i hope you are fine in other aspects of your life as well
(i lurk silently like a ninja
)
stardustsailor
Well-Known Member
No Pet ...
There's no IC ,that is going to replace a 'led driver' the way ,I've understood ,that you mean ....
One can use those IC's ,along with few other components in order to make a CC led driver ...
But sheldomly those ICs can handle high output power and if so ,their price is neither 'forgiving' in case of a
'screw-up' ,neither has a big difference with a ready-made CC driver of same power range ...
Anyway ...
Probably,for those who either want something really 'custom' or either have plenty time/money to give away/lose ...
(Regarding the DIYers ,not the manufacturers ...)
