MaxFlex on LED heatsink??

Candle Power Forums

Help Support Candle Power:

piesoup

Newly Enlightened
Joined
Oct 31, 2008
Messages
114
City & State/Province
East Anglia
Hello
I am wondering if it is feasable to mount my MaxFlex on the rear of the heatsink for my LEDs? I dont want the heat from the LEDs to heat up the driver as opposed to it drawing the heat out if it!
* I'll have 4x XR-Es @1000mA
* Mounted on a 53mm dia X 15mm deep solid aluminium heatsink.
* Heatsink will have a snug fit inside a 50mm long aluminuim machined cylinder with a 57mm OD and 53mm ID.

Or will it be better to have a seperate thermal path to the light body for the driver. :thinking:
Sorry if it sounds like a dumb *** question, I waited ages for my driver to arrive from cutters and i dont want to fry it!
Cheers peeps
 
There has been some debate about this and truthfully either approach should work. Personally, I like to mount the maxFlex close to the LEDs so that the thermal protection capability can be used to monitor and protect the LEDs. I usually set the maxFlex thermal protection for 60C or 70C, if your heatsink gets much hotter than that the LEDs are probably getting close to the limit anyways (especially if they're mounted on stars). The maxFlex can easily cope with running at that temp, so as long as you make a good thermal connection to the housing and enable the thermal protection it will be fine no matter where you mount it.
 
If you are not pushing the maxflex too hard it should be ok and is good as the thermal protection protects your leds. With an 11.1v 3 cell setup OK I would say OK. If you are pushing the maxflex hard e.g. a 7.2v 2 cell setup then you are probobly better with seperate heatsinking for the maxflex. I have had issues with the latest maxflexes overheating the switcher chip and it shutting off. If you are pushing it hard eg > 2A in it can heat up quicker than the thermal protection has a chance to do its job. The TaskLed maxflex documentation has been updated to reflect this issue recently.

Ifor
 
Input V is 12v NiMh
Input current will be 1.9A
Power out will be 16.8W
With losses in the driver at about 1.512W, depending on how efficient it is.
From that I will need some pretty good heatsinking! I'm going to try and work out a way of getting rid of some of the heat from the top of the switcher IC, U1. I had forgotten about the heat protection in the driver, using it to protect the LEDs!
Thanks for the replies guys
 
I AA'd a strip of copper plate to the top of the switcher ic on a blue shark,which was soldered to the sharks heatsink. Seems to draw the heat away very well but it's a bit labour intensive.
 
Input V is 12v NiMh
Input current will be 1.9A
Power out will be 16.8W
With losses in the driver at about 1.512W, depending on how efficient it is.
From that I will need some pretty good heatsinking! I'm going to try and work out a way of getting rid of some of the heat from the top of the switcher IC, U1. I had forgotten about the heat protection in the driver, using it to protect the LEDs!
Thanks for the replies guys

Not sure where your 16.8W power out is coming from. With 1000mA to 4 LEDs and assuming a Vf around 3.7V you have 4 x 3.7 x 1 = 14.8W. You have a total Vf of 14.8V (at 1A) and 12V in, driver efficiency will be very good and likely ~93%.

So, you then have 14.8W / 12V / 0.93 = 1.3A input.

Obviously these are you best case numbers while your battery pack is fresh at 12V. But fully discharged they are much worse (10V total battery pack voltage). Make sure you set up the voltage monitoring thresholds in the menu for you battery pack (by default it is configured for 2 x li-ion cells).

With efficiency around 93% and 14.8W output, heat losses will be <1.2W. Heatsinking is needed and mounting it to the gold area of the PCB (UNDER the switcher IC) will work just fine. You don't need to mount/glue anything to the top of the IC package.

I assume you've read the thermal calculations at www.taskled.com/techmaxflex.html

cheers,
george.
 
Hi George
Thanks for your reply. I have read the .PDF on your site, very useful! The reason I got a different p out as because I used the worst case for the driver, ie. 10v as the lowest voltage from my battery and 1200mA in case I ran it on max. At least that's what I think I did! I think I used an effy of 90% too.
So for normal running on a good battery it'll be ok plonking the driver on the back of the heatsink. Phew! I've just finished machiningthe casing and it looks pretty chunky!
 

Latest posts

Back
Top