Running two 3.7V LEDs from 12V

That's what the driver does. It's a buck regulator (try wikipedia) that controls itself based on the current passing through it. It's a bit more complicated than this, but think of it as a switch that turns itself on and off very rapidly. The duty cycle (ratio of on-time to off-time) is controlled by some feedback loop.

In the case of a current regulator, the input to that control loop is the voltage developed across a low-value sense resistor, which is located on the driver board. This resistor is usually in the negative (return) lead from the load. For this reason, LED - is probably not common to the negative input, and you should keep this in mind in your design.

Please post back if you have further questions, we were all new once, and wrapping your head around proper use of regulated supplies is important!
 
I've head that the voltages in a car can jump around a lot so I'd recommend a more robust circuit then that. One that can handle surges of voltage.

If the voltage was relatively constant then the driver will drop the voltage to what the LED needs(that's what a buck driver does).
 
Thank you both for your helpful replies. It is so nice people give decent advice to newbies.

I just tested the LED and driver and have a big grin on my face now. The completed lights are to be used as reverse lights in my car as the stock ones are poor.

I had some neat little heatsinks made for the project:

_MG_6812.jpg


_MG_6809.jpg


The final heatsinks will accomodate two LEDs to give plenty of output. I have heard automotive 12V can be very spikey so was considering using a zener diode to prevent any peaks. I will look into that further down the line. Since the lights are not on all the time and will not be on when the engine is started/stopped I think the risk is reduced.

The heatsink got warm but stayed cooler than I was expecting. The LED was held on with some thermal compound and little blobs of cyanoacrylate adhesive at the edge.

Once again, thanks for taking the time to help me out in my project :thumbsup:
Dave
 
I hope it works well!

It think you should get a zener diode just in case but the driver might end up being cheaper then the diode.
 
Once again thanks for your kind words of encouragement.

The first aluminium heatsinks pictured were made wrong but I was using them for testing. My new ones arrived today from A Wrate Engineering www.awrate.co.uk:

_MG_6817.jpg


There will be one star board mounted on each side. I just need to pick up a metal file on the way back from work tomorrow to shave the stars down a tad to fit on the heatsinks.

Really getting exited about this now. In last nights test one LED was brighter than one stock bulb, so the total of 4 will be at least 4 times brighter :D
 
After plenty of searching I found a thread where someone had placed the heated star on a metal plate to remove the emitter.

I simply put the star board on the aga plate. Someone used a knife to hold down the board whilst I pushed the emitter sideways. As soon as it moved I lifted one edge up and picked it up with some tweezers. It went straight onto a metal surface to conduct away the heat. I just hooked it up and it seems to be working fine.

I will now mount the two bare emitters to the heatsinks using some Arctic Silver and cyanoacrylate adhesive at the edges. It looks a lot neater and should give better thermal coupling to the heatsink :D
 
Top